id
stringlengths
11
11
created
timestamp[s]date
2026-01-01 00:00:00
2026-01-01 00:00:00
topic
stringclasses
14 values
task_type
stringclasses
10 values
difficulty
stringclasses
3 values
instruction
stringlengths
189
248
input
stringclasses
1 value
output
stringclasses
9 values
reasoning_steps
listlengths
0
5
metadata
dict
hash
stringlengths
40
40
train_20700
2026-01-01T00:00:00
Extended context and repo-scale understanding
patch_diff
expert
Task: patch_diff Topic: Extended context and repo-scale understanding Difficulty: expert Target language: Go Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[]
{ "target_language": "Go", "developer_needs": [ "ci_integration", "tests_are_truth", "reproducibility", "auditability" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
5b15f92a432768e5b914d71782e442d4459686db
train_20701
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
code
intermediate
Task: code Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: intermediate Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Scaffold: ```python def agent_loop(plan, edit, test, issue, max_iters=4): history = [] p = plan(issue) for _ in range(max_iters): patch = edit(issue, p) ok, report = test(patch) history.append({"plan": p, "ok": ok}) if ok: return patch, history p = p + " | refine" return patch, history ```
[]
{ "target_language": "Python", "developer_needs": [ "reproducibility", "governance", "cost_latency_tradeoffs", "tests_are_truth" ], "moe_experts": [ "coding_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
672e9f4a5f927846dc1669585d9676175bc8aebf
train_20702
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
design
intermediate
Task: design Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: intermediate Target language: TypeScript Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "TypeScript", "developer_needs": [ "tooling", "repo_scale_reasoning", "auditability", "documentation" ], "moe_experts": [ "security_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4aa2a38ee97d1b1519e44f2004503d1a588f5e73
train_20703
2026-01-01T00:00:00
Latency, cost, and reliability optimization
code
expert
Task: code Topic: Latency, cost, and reliability optimization Difficulty: expert Target language: TypeScript Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "documentation", "cost_latency_tradeoffs", "evaluation_metrics", "auditability" ], "moe_experts": [ "agentic_systems_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
915640a0bfe795fe4da66a8359f3dcb2888ccbe0
train_20704
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
data_pipeline
advanced
Task: data_pipeline Topic: SWE-bench style real-repo evaluation Difficulty: advanced Target language: TypeScript Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "tests_are_truth", "documentation", "reproducibility", "security_gates" ], "moe_experts": [ "governance_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
026ca1c8ee3cea588df4dbd1cb8cc25e13743f76
train_20705
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
agent_loop
expert
Task: agent_loop Topic: Mixture-of-Experts (MoE) for code Difficulty: expert Target language: JavaScript Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "JavaScript", "developer_needs": [ "tooling", "evaluation_metrics", "tests_are_truth", "auditability" ], "moe_experts": [ "governance_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9978bb0e05d6a2eca91ae94ba5bd1e24329a1db4
train_20706
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
data_pipeline
intermediate
Task: data_pipeline Topic: Tool calling, sandboxes, and CI integration Difficulty: intermediate Target language: TypeScript Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[]
{ "target_language": "TypeScript", "developer_needs": [ "ci_integration", "reproducibility", "documentation", "auditability" ], "moe_experts": [ "security_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6accc5c864f1377519675abea3537f7eb645ffd5
train_20707
2026-01-01T00:00:00
Self-improving agents and feedback loops
code
advanced
Task: code Topic: Self-improving agents and feedback loops Difficulty: advanced Target language: C# Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "C#", "developer_needs": [ "repo_scale_reasoning", "ci_integration", "tooling", "security_gates" ], "moe_experts": [ "performance_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d66db1a609395cf1dac7441afeb1c8a4cae2344a
train_20708
2026-01-01T00:00:00
Dataset curation pipelines (filter, dedupe, quality)
patch_diff
expert
Task: patch_diff Topic: Dataset curation pipelines (filter, dedupe, quality) Difficulty: expert Target language: Bash Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Bash", "developer_needs": [ "auditability", "cost_latency_tradeoffs", "evaluation_metrics", "tooling" ], "moe_experts": [ "evaluation_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d2fd24050e464465299066212e243c4fd729b0dd
train_20709
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
explain
advanced
Task: explain Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: advanced Target language: JavaScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "JavaScript", "developer_needs": [ "security_gates", "governance", "reproducibility", "documentation" ], "moe_experts": [ "performance_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
c03ca3603a8078f5daee7f65c6b1a4c065d4057a
train_20710
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
review
intermediate
Task: review Topic: Tool calling, sandboxes, and CI integration Difficulty: intermediate Target language: Go Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "Go", "developer_needs": [ "governance", "reproducibility", "auditability", "repo_scale_reasoning" ], "moe_experts": [ "evaluation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
16f5aefbbc2768ad0e36a5d53299d78bae2d8d68
train_20711
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
review
expert
Task: review Topic: Tool calling, sandboxes, and CI integration Difficulty: expert Target language: TypeScript Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "security_gates", "evaluation_metrics", "tooling", "ci_integration" ], "moe_experts": [ "agentic_systems_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6c5e7ba70394951afd8a1e86d824a520f4d2dc80
train_20712
2026-01-01T00:00:00
Self-improving agents and feedback loops
failure_analysis
expert
Task: failure_analysis Topic: Self-improving agents and feedback loops Difficulty: expert Target language: Bash Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Failure: - Initial patch broke edge case Reflection: - Missing zero-input guard Correction: - Add explicit validation + test
[]
{ "target_language": "Bash", "developer_needs": [ "ci_integration", "governance", "tooling", "tests_are_truth" ], "moe_experts": [ "evaluation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d557e21fe235028d0e707cdf3ae8c17e1bc7feb5
train_20713
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
patch_diff
intermediate
Task: patch_diff Topic: Reasoning-first coding models and tunable deliberation Difficulty: intermediate Target language: Java Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[]
{ "target_language": "Java", "developer_needs": [ "governance", "ci_integration", "tooling", "auditability" ], "moe_experts": [ "data_curation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
71332c40d6188c096470ef6b5224c87abf71c6c1
train_20714
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
review
advanced
Task: review Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: advanced Target language: Bash Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "Bash", "developer_needs": [ "repo_scale_reasoning", "ci_integration", "reproducibility", "tests_are_truth" ], "moe_experts": [ "agentic_systems_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
65708bebf748ca275fabb103aa632ca480830b5a
train_20715
2026-01-01T00:00:00
Extended context and repo-scale understanding
eval
intermediate
Task: eval Topic: Extended context and repo-scale understanding Difficulty: intermediate Target language: C# Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[]
{ "target_language": "C#", "developer_needs": [ "cost_latency_tradeoffs", "documentation", "tests_are_truth", "repo_scale_reasoning" ], "moe_experts": [ "performance_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9ed7e24fbf054de36059c3e7fc06ae443fdb70b8
train_20716
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
code
expert
Task: code Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: expert Target language: TypeScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "TypeScript", "developer_needs": [ "governance", "tooling", "cost_latency_tradeoffs", "evaluation_metrics" ], "moe_experts": [ "governance_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
64eeeadb6d0b6854cf0e28c351d4ead875d98049
train_20717
2026-01-01T00:00:00
Secure code generation and policy gates
compare
intermediate
Task: compare Topic: Secure code generation and policy gates Difficulty: intermediate Target language: Python Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[]
{ "target_language": "Python", "developer_needs": [ "security_gates", "cost_latency_tradeoffs", "tests_are_truth", "evaluation_metrics" ], "moe_experts": [ "agentic_systems_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
76955ffb2f252c7655e2722d95fa72137b1be531
train_20718
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
compare
advanced
Task: compare Topic: Governance, provenance, and licensing for code data Difficulty: advanced Target language: Rust Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "tests_are_truth", "evaluation_metrics", "documentation", "reproducibility" ], "moe_experts": [ "governance_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
3c95b593fc1d6dcafcdb6740e8c9f95669cda67f
train_20719
2026-01-01T00:00:00
Secure code generation and policy gates
patch_diff
advanced
Task: patch_diff Topic: Secure code generation and policy gates Difficulty: advanced Target language: Rust Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[]
{ "target_language": "Rust", "developer_needs": [ "security_gates", "governance", "evaluation_metrics", "auditability" ], "moe_experts": [ "data_curation_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
3c6b6324552e978b798e26912a951bcbbcd44552
train_20720
2026-01-01T00:00:00
Secure code generation and policy gates
compare
expert
Task: compare Topic: Secure code generation and policy gates Difficulty: expert Target language: Java Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Java", "developer_needs": [ "security_gates", "governance", "auditability", "ci_integration" ], "moe_experts": [ "agentic_systems_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
f56bed9b7ce6e49092da83a6570d63af613bd8b8
train_20721
2026-01-01T00:00:00
Self-improving agents and feedback loops
failure_analysis
intermediate
Task: failure_analysis Topic: Self-improving agents and feedback loops Difficulty: intermediate Target language: Go Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Failure: - Initial patch broke edge case Reflection: - Missing zero-input guard Correction: - Add explicit validation + test
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "tooling", "repo_scale_reasoning", "evaluation_metrics", "documentation" ], "moe_experts": [ "security_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
617d5e9644b5d5f58e9bd88655b253f0148fe3f3
train_20722
2026-01-01T00:00:00
Multimodal dev workflows (docs, diagrams, traces)
review
expert
Task: review Topic: Multimodal dev workflows (docs, diagrams, traces) Difficulty: expert Target language: TypeScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "TypeScript", "developer_needs": [ "governance", "repo_scale_reasoning", "reproducibility", "security_gates" ], "moe_experts": [ "evaluation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4b198e67a17425268624cf60538c68886f73113e
train_20723
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
eval
expert
Task: eval Topic: Governance, provenance, and licensing for code data Difficulty: expert Target language: Go Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "tests_are_truth", "tooling", "repo_scale_reasoning", "cost_latency_tradeoffs" ], "moe_experts": [ "governance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9d8d5d1ee717533064587232cadd72f1ab29cda4
train_20724
2026-01-01T00:00:00
Self-improving agents and feedback loops
patch_diff
intermediate
Task: patch_diff Topic: Self-improving agents and feedback loops Difficulty: intermediate Target language: Rust Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "reproducibility", "repo_scale_reasoning", "tooling", "tests_are_truth" ], "moe_experts": [ "coding_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
223e7781d4a7af911a27617028ea73b1840e1aaa
train_20725
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
explain
advanced
Task: explain Topic: SWE-bench style real-repo evaluation Difficulty: advanced Target language: SQL Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "SQL", "developer_needs": [ "documentation", "repo_scale_reasoning", "tooling", "ci_integration" ], "moe_experts": [ "performance_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
37ec1f1b6a44e3e47cbbaad5c91d8270dd0d9fde
train_20726
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
explain
advanced
Task: explain Topic: SWE-bench style real-repo evaluation Difficulty: advanced Target language: Go Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Go", "developer_needs": [ "tests_are_truth", "tooling", "security_gates", "evaluation_metrics" ], "moe_experts": [ "coding_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4628b463ca6a385509cc99acd87649f9401b3063
train_20727
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
compare
advanced
Task: compare Topic: Model merging, distillation, and continued pretraining Difficulty: advanced Target language: Java Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Java", "developer_needs": [ "cost_latency_tradeoffs", "repo_scale_reasoning", "security_gates", "ci_integration" ], "moe_experts": [ "governance_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
88261a85b059501f759883f82c7444973ac258fd
train_20728
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
patch_diff
intermediate
Task: patch_diff Topic: SWE-bench style real-repo evaluation Difficulty: intermediate Target language: Go Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "documentation", "tooling", "tests_are_truth", "reproducibility" ], "moe_experts": [ "performance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
856cd150b84836264469c2704defe8ba172bc2f0
train_20729
2026-01-01T00:00:00
Self-improving agents and feedback loops
failure_analysis
expert
Task: failure_analysis Topic: Self-improving agents and feedback loops Difficulty: expert Target language: TypeScript Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Failure: - Initial patch broke edge case Reflection: - Missing zero-input guard Correction: - Add explicit validation + test
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "tooling", "cost_latency_tradeoffs", "security_gates", "auditability" ], "moe_experts": [ "data_curation_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d1371523d0e3607b7a873aa3b3ca60c3187ea7e2
train_20730
2026-01-01T00:00:00
Secure code generation and policy gates
compare
intermediate
Task: compare Topic: Secure code generation and policy gates Difficulty: intermediate Target language: SQL Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "SQL", "developer_needs": [ "repo_scale_reasoning", "security_gates", "cost_latency_tradeoffs", "tooling" ], "moe_experts": [ "performance_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
15c103ba376755d289d36fc591c533f4dcc1bc49
train_20731
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
eval
advanced
Task: eval Topic: Tool calling, sandboxes, and CI integration Difficulty: advanced Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Python", "developer_needs": [ "auditability", "governance", "cost_latency_tradeoffs", "tests_are_truth" ], "moe_experts": [ "evaluation_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
ae556a09e86d9170118acc22dd349363428fc783
train_20732
2026-01-01T00:00:00
Code-specialized model families and sizing tradeoffs
explain
advanced
Task: explain Topic: Code-specialized model families and sizing tradeoffs Difficulty: advanced Target language: TypeScript Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "TypeScript", "developer_needs": [ "reproducibility", "security_gates", "cost_latency_tradeoffs", "tests_are_truth" ], "moe_experts": [ "governance_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
3e9c29839dfd37aa3678d4a67ab5116f4a53a2e9
train_20733
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
explain
intermediate
Task: explain Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: intermediate Target language: Python Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Python", "developer_needs": [ "tests_are_truth", "cost_latency_tradeoffs", "documentation", "repo_scale_reasoning" ], "moe_experts": [ "performance_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
c0593b036da998ceb3cdf6f38229f3550b6522db
train_20734
2026-01-01T00:00:00
Latency, cost, and reliability optimization
eval
intermediate
Task: eval Topic: Latency, cost, and reliability optimization Difficulty: intermediate Target language: JavaScript Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "JavaScript", "developer_needs": [ "reproducibility", "documentation", "cost_latency_tradeoffs", "ci_integration" ], "moe_experts": [ "performance_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
489b4210a05cf21177fb343ba05b038c1aabec7c
train_20735
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
data_pipeline
expert
Task: data_pipeline Topic: SWE-bench style real-repo evaluation Difficulty: expert Target language: Java Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Java", "developer_needs": [ "documentation", "ci_integration", "repo_scale_reasoning", "tests_are_truth" ], "moe_experts": [ "governance_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
221d802af38eb1cd86fce9f34f66275c22c32a80
train_20736
2026-01-01T00:00:00
Dataset curation pipelines (filter, dedupe, quality)
data_pipeline
intermediate
Task: data_pipeline Topic: Dataset curation pipelines (filter, dedupe, quality) Difficulty: intermediate Target language: SQL Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "SQL", "developer_needs": [ "tooling", "reproducibility", "repo_scale_reasoning", "security_gates" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
721731e2b9160beb4f329626808cd4fc88afccd5
train_20737
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
data_pipeline
intermediate
Task: data_pipeline Topic: Reasoning-first coding models and tunable deliberation Difficulty: intermediate Target language: Go Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "auditability", "governance", "security_gates", "reproducibility" ], "moe_experts": [ "evaluation_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
7f38d82bbfc56caac17f73a7d775796a1fc6e9c1
train_20738
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
eval
advanced
Task: eval Topic: Reasoning-first coding models and tunable deliberation Difficulty: advanced Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Python", "developer_needs": [ "ci_integration", "tooling", "tests_are_truth", "auditability" ], "moe_experts": [ "performance_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
78111ccd211f16c6002b1547cade30c5a97ce9dd
train_20739
2026-01-01T00:00:00
Multimodal dev workflows (docs, diagrams, traces)
review
expert
Task: review Topic: Multimodal dev workflows (docs, diagrams, traces) Difficulty: expert Target language: JavaScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "JavaScript", "developer_needs": [ "tests_are_truth", "security_gates", "documentation", "repo_scale_reasoning" ], "moe_experts": [ "coding_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9b9a014361e4607907d9fcb60419a031d7ae6dd5
train_20740
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
data_pipeline
intermediate
Task: data_pipeline Topic: Tool calling, sandboxes, and CI integration Difficulty: intermediate Target language: C# Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "C#", "developer_needs": [ "evaluation_metrics", "governance", "tooling", "security_gates" ], "moe_experts": [ "governance_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9c86a6c2e19bbc86e9711fb63593394cf29eb27c
train_20741
2026-01-01T00:00:00
Secure code generation and policy gates
code
intermediate
Task: code Topic: Secure code generation and policy gates Difficulty: intermediate Target language: Go Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Go", "developer_needs": [ "auditability", "security_gates", "documentation", "cost_latency_tradeoffs" ], "moe_experts": [ "evaluation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
ff59803c2eca85123f794a75cdfdfde3d586dcee
train_20742
2026-01-01T00:00:00
Multimodal dev workflows (docs, diagrams, traces)
review
intermediate
Task: review Topic: Multimodal dev workflows (docs, diagrams, traces) Difficulty: intermediate Target language: Rust Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "Rust", "developer_needs": [ "auditability", "evaluation_metrics", "ci_integration", "reproducibility" ], "moe_experts": [ "security_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
190303f9d6dd3327407fe1fcd8cd9d2b20a66998
train_20743
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
eval
expert
Task: eval Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: expert Target language: Go Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "reproducibility", "evaluation_metrics", "governance", "documentation" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
0d804f9abe36074cda88a1465141af707496f524
train_20744
2026-01-01T00:00:00
Extended context and repo-scale understanding
design
expert
Task: design Topic: Extended context and repo-scale understanding Difficulty: expert Target language: Java Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Java", "developer_needs": [ "governance", "tooling", "reproducibility", "ci_integration" ], "moe_experts": [ "evaluation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
c1efd76f690c7ef7de67e9b20387973f3696ec9d
train_20745
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
explain
expert
Task: explain Topic: Mixture-of-Experts (MoE) for code Difficulty: expert Target language: SQL Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "SQL", "developer_needs": [ "tooling", "auditability", "governance", "cost_latency_tradeoffs" ], "moe_experts": [ "coding_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
5265a0bc44fc1b12b8fcb8d2063f30518bbdd83a
train_20746
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
agent_loop
expert
Task: agent_loop Topic: Mixture-of-Experts (MoE) for code Difficulty: expert Target language: TypeScript Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[]
{ "target_language": "TypeScript", "developer_needs": [ "evaluation_metrics", "auditability", "reproducibility", "governance" ], "moe_experts": [ "agentic_systems_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
630e0cf864bfb93dbc3491dedb0319742a6fba8a
train_20747
2026-01-01T00:00:00
Extended context and repo-scale understanding
explain
intermediate
Task: explain Topic: Extended context and repo-scale understanding Difficulty: intermediate Target language: Rust Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "governance", "documentation", "auditability", "evaluation_metrics" ], "moe_experts": [ "evaluation_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
a62ef85c5d859b0083d1b12dceaa09a916960d3f
train_20748
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
compare
advanced
Task: compare Topic: Reasoning-first coding models and tunable deliberation Difficulty: advanced Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "reproducibility", "tests_are_truth", "repo_scale_reasoning", "tooling" ], "moe_experts": [ "agentic_systems_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d11301a105adc727e13935aa8ff11a856c0b63f5
train_20749
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
review
advanced
Task: review Topic: Governance, provenance, and licensing for code data Difficulty: advanced Target language: SQL Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "SQL", "developer_needs": [ "cost_latency_tradeoffs", "governance", "security_gates", "documentation" ], "moe_experts": [ "governance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4d1ea854fe6e39005e7c249987a163a5aa8e436e
train_20750
2026-01-01T00:00:00
Code-specialized model families and sizing tradeoffs
code
advanced
Task: code Topic: Code-specialized model families and sizing tradeoffs Difficulty: advanced Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Rust", "developer_needs": [ "evaluation_metrics", "ci_integration", "security_gates", "tooling" ], "moe_experts": [ "governance_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
bba8c1b2375baa79b25d8b68a8416c88ae0a1a5d
train_20751
2026-01-01T00:00:00
Extended context and repo-scale understanding
patch_diff
advanced
Task: patch_diff Topic: Extended context and repo-scale understanding Difficulty: advanced Target language: Rust Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "documentation", "cost_latency_tradeoffs", "security_gates", "governance" ], "moe_experts": [ "data_curation_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
90175149607c89c0c20c0088e9c22577d6573af6
train_20752
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
code
advanced
Task: code Topic: Governance, provenance, and licensing for code data Difficulty: advanced Target language: Rust Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "evaluation_metrics", "security_gates", "ci_integration", "repo_scale_reasoning" ], "moe_experts": [ "agentic_systems_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d9ce05fda1f0bd8d87b1ad59016d583307124232
train_20753
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
agent_loop
expert
Task: agent_loop Topic: Tool calling, sandboxes, and CI integration Difficulty: expert Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[]
{ "target_language": "Rust", "developer_needs": [ "ci_integration", "governance", "evaluation_metrics", "tooling" ], "moe_experts": [ "evaluation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
b9088aafa87b7814a754d27d020f7254e42d0864
train_20754
2026-01-01T00:00:00
Latency, cost, and reliability optimization
eval
intermediate
Task: eval Topic: Latency, cost, and reliability optimization Difficulty: intermediate Target language: Bash Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[]
{ "target_language": "Bash", "developer_needs": [ "ci_integration", "tests_are_truth", "tooling", "evaluation_metrics" ], "moe_experts": [ "evaluation_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6700303413640188721295bc1899d12ee5aeb954
train_20755
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
patch_diff
advanced
Task: patch_diff Topic: SWE-bench style real-repo evaluation Difficulty: advanced Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[]
{ "target_language": "Python", "developer_needs": [ "evaluation_metrics", "tests_are_truth", "ci_integration", "security_gates" ], "moe_experts": [ "coding_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
27574483c0e3f5f3c221a07971897e65bd207d47
train_20756
2026-01-01T00:00:00
Latency, cost, and reliability optimization
code
advanced
Task: code Topic: Latency, cost, and reliability optimization Difficulty: advanced Target language: Bash Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Bash", "developer_needs": [ "tests_are_truth", "auditability", "evaluation_metrics", "governance" ], "moe_experts": [ "security_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
979181b56c409a88dfc9cbce96fd4a9c34d3a6c8
train_20757
2026-01-01T00:00:00
Self-improving agents and feedback loops
explain
advanced
Task: explain Topic: Self-improving agents and feedback loops Difficulty: advanced Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "documentation", "cost_latency_tradeoffs", "tests_are_truth", "reproducibility" ], "moe_experts": [ "data_curation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
0fc249e1da72bec3cd6dde33f6856b19e0d9ba17
train_20758
2026-01-01T00:00:00
Dataset curation pipelines (filter, dedupe, quality)
review
expert
Task: review Topic: Dataset curation pipelines (filter, dedupe, quality) Difficulty: expert Target language: TypeScript Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "repo_scale_reasoning", "tooling", "security_gates", "governance" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
e3076447d6116dbae7efa569ee71e262631f2ebc
train_20759
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
explain
advanced
Task: explain Topic: Reasoning-first coding models and tunable deliberation Difficulty: advanced Target language: Java Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Java", "developer_needs": [ "evaluation_metrics", "ci_integration", "reproducibility", "cost_latency_tradeoffs" ], "moe_experts": [ "agentic_systems_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
bb64f43f805e0d7640f4ffe9a0c5338ee60bd8fd
train_20760
2026-01-01T00:00:00
Extended context and repo-scale understanding
code
advanced
Task: code Topic: Extended context and repo-scale understanding Difficulty: advanced Target language: Go Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Go", "developer_needs": [ "documentation", "cost_latency_tradeoffs", "evaluation_metrics", "governance" ], "moe_experts": [ "coding_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
7afdf89492526efbbcbefe940760b5d9b54c36e8
train_20761
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
data_pipeline
intermediate
Task: data_pipeline Topic: Governance, provenance, and licensing for code data Difficulty: intermediate Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[]
{ "target_language": "Python", "developer_needs": [ "auditability", "tooling", "repo_scale_reasoning", "ci_integration" ], "moe_experts": [ "security_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
ec023554f72bc7a11533baaec74f01f76c721be7
train_20762
2026-01-01T00:00:00
Code-specialized model families and sizing tradeoffs
design
intermediate
Task: design Topic: Code-specialized model families and sizing tradeoffs Difficulty: intermediate Target language: Go Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Go", "developer_needs": [ "tooling", "ci_integration", "tests_are_truth", "reproducibility" ], "moe_experts": [ "coding_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
c41419c5eba8fa123e46e4e6524c739b7fc6a555
train_20763
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
patch_diff
intermediate
Task: patch_diff Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: intermediate Target language: C# Context: Regulated environment requiring audit trails. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "C#", "developer_needs": [ "governance", "repo_scale_reasoning", "reproducibility", "tests_are_truth" ], "moe_experts": [ "data_curation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
8e9fc70edb7b7ad72701071a30d796221dc9d443
train_20764
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
patch_diff
intermediate
Task: patch_diff Topic: Mixture-of-Experts (MoE) for code Difficulty: intermediate Target language: Go Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[]
{ "target_language": "Go", "developer_needs": [ "documentation", "security_gates", "governance", "cost_latency_tradeoffs" ], "moe_experts": [ "coding_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9650bf9e318968463320a78eac5afe00dfcb6b60
train_20765
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
failure_analysis
expert
Task: failure_analysis Topic: Mixture-of-Experts (MoE) for code Difficulty: expert Target language: Rust Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Failure: - Initial patch broke edge case Reflection: - Missing zero-input guard Correction: - Add explicit validation + test
[]
{ "target_language": "Rust", "developer_needs": [ "ci_integration", "tooling", "security_gates", "auditability" ], "moe_experts": [ "security_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
b7de1d117a9c9b55e37bb36360572d5d0fee614a
train_20766
2026-01-01T00:00:00
Dataset curation pipelines (filter, dedupe, quality)
agent_loop
intermediate
Task: agent_loop Topic: Dataset curation pipelines (filter, dedupe, quality) Difficulty: intermediate Target language: Java Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[]
{ "target_language": "Java", "developer_needs": [ "reproducibility", "evaluation_metrics", "ci_integration", "documentation" ], "moe_experts": [ "coding_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
f0b94d00628cf9e82d665ed961e9e32a8f33e85a
train_20767
2026-01-01T00:00:00
Mixture-of-Experts (MoE) for code
data_pipeline
expert
Task: data_pipeline Topic: Mixture-of-Experts (MoE) for code Difficulty: expert Target language: C# Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[]
{ "target_language": "C#", "developer_needs": [ "repo_scale_reasoning", "security_gates", "reproducibility", "tooling" ], "moe_experts": [ "data_curation_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
7e2341e2df02f7143282db0562a37bdd2f2f5bde
train_20768
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
code
intermediate
Task: code Topic: Model merging, distillation, and continued pretraining Difficulty: intermediate Target language: Bash Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Bash", "developer_needs": [ "evaluation_metrics", "ci_integration", "cost_latency_tradeoffs", "auditability" ], "moe_experts": [ "security_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6ff2dad90e1e1276d3ddc93b22a603c11fc33cf8
train_20769
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
eval
expert
Task: eval Topic: SWE-bench style real-repo evaluation Difficulty: expert Target language: JavaScript Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "JavaScript", "developer_needs": [ "repo_scale_reasoning", "cost_latency_tradeoffs", "ci_integration", "security_gates" ], "moe_experts": [ "governance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
b3d02dfc330a739aa56e7d438c43ca3ee9113e81
train_20770
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
design
intermediate
Task: design Topic: Governance, provenance, and licensing for code data Difficulty: intermediate Target language: JavaScript Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "JavaScript", "developer_needs": [ "ci_integration", "tooling", "cost_latency_tradeoffs", "tests_are_truth" ], "moe_experts": [ "evaluation_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
496d95f76d8a4412af0d1f426e8a7b3d08c7e722
train_20771
2026-01-01T00:00:00
Multimodal dev workflows (docs, diagrams, traces)
compare
expert
Task: compare Topic: Multimodal dev workflows (docs, diagrams, traces) Difficulty: expert Target language: Bash Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Bash", "developer_needs": [ "reproducibility", "ci_integration", "tooling", "governance" ], "moe_experts": [ "performance_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
53288cc75a99698bb12cbc0c336bd67c3a896886
train_20772
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
design
expert
Task: design Topic: Tool calling, sandboxes, and CI integration Difficulty: expert Target language: Bash Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Bash", "developer_needs": [ "documentation", "ci_integration", "tooling", "governance" ], "moe_experts": [ "governance_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
324b40e4020a412b2da0d8731b0b107977d302fb
train_20773
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
review
expert
Task: review Topic: Model merging, distillation, and continued pretraining Difficulty: expert Target language: SQL Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "SQL", "developer_needs": [ "ci_integration", "governance", "tests_are_truth", "evaluation_metrics" ], "moe_experts": [ "agentic_systems_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
56d68f00b629e6d8087b9968114f190060e2e437
train_20774
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
design
intermediate
Task: design Topic: Model merging, distillation, and continued pretraining Difficulty: intermediate Target language: Python Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Python", "developer_needs": [ "documentation", "evaluation_metrics", "security_gates", "tooling" ], "moe_experts": [ "coding_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6a819d7d2345aea3d5e6d8c08516baf6095763f5
train_20775
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
patch_diff
advanced
Task: patch_diff Topic: SWE-bench style real-repo evaluation Difficulty: advanced Target language: Java Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Java", "developer_needs": [ "evaluation_metrics", "ci_integration", "documentation", "reproducibility" ], "moe_experts": [ "evaluation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
6757200439f8f444c86e534cb64b75ff62e0c358
train_20776
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
review
intermediate
Task: review Topic: Governance, provenance, and licensing for code data Difficulty: intermediate Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "repo_scale_reasoning", "evaluation_metrics", "reproducibility", "cost_latency_tradeoffs" ], "moe_experts": [ "coding_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
21710c2e5e2bfa14ac76ab06ac8ccd880539c3d6
train_20777
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
agent_loop
advanced
Task: agent_loop Topic: Model merging, distillation, and continued pretraining Difficulty: advanced Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "auditability", "cost_latency_tradeoffs", "governance", "tooling" ], "moe_experts": [ "evaluation_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
9afa47e8f9962338525d51555d2e45659a124bb8
train_20778
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
data_pipeline
expert
Task: data_pipeline Topic: Governance, provenance, and licensing for code data Difficulty: expert Target language: JavaScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[]
{ "target_language": "JavaScript", "developer_needs": [ "repo_scale_reasoning", "auditability", "documentation", "security_gates" ], "moe_experts": [ "evaluation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
10ed97d645168d440dfbd9e2a95a6808df94dc87
train_20779
2026-01-01T00:00:00
Dataset curation pipelines (filter, dedupe, quality)
code
expert
Task: code Topic: Dataset curation pipelines (filter, dedupe, quality) Difficulty: expert Target language: SQL Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "SQL", "developer_needs": [ "cost_latency_tradeoffs", "repo_scale_reasoning", "evaluation_metrics", "security_gates" ], "moe_experts": [ "security_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
3f610c852a4140d0237cda85534561b0f7ab8e03
train_20780
2026-01-01T00:00:00
Extended context and repo-scale understanding
explain
expert
Task: explain Topic: Extended context and repo-scale understanding Difficulty: expert Target language: Bash Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Bash", "developer_needs": [ "cost_latency_tradeoffs", "documentation", "security_gates", "evaluation_metrics" ], "moe_experts": [ "evaluation_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
982f41159f39bf4fd113f8ff143a02353a639893
train_20781
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
explain
intermediate
Task: explain Topic: SWE-bench style real-repo evaluation Difficulty: intermediate Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Rust", "developer_needs": [ "cost_latency_tradeoffs", "repo_scale_reasoning", "auditability", "documentation" ], "moe_experts": [ "security_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
46628ac18c7bbadf99b31edbac3ce6781a96d4a2
train_20782
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
data_pipeline
advanced
Task: data_pipeline Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: advanced Target language: C# Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[]
{ "target_language": "C#", "developer_needs": [ "tooling", "repo_scale_reasoning", "auditability", "reproducibility" ], "moe_experts": [ "governance_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
dfde98547c62e897353db2ce00e037edadab8238
train_20783
2026-01-01T00:00:00
Code-specialized model families and sizing tradeoffs
code
advanced
Task: code Topic: Code-specialized model families and sizing tradeoffs Difficulty: advanced Target language: Java Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Java", "developer_needs": [ "cost_latency_tradeoffs", "tooling", "governance", "documentation" ], "moe_experts": [ "performance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
c4dafbb0b8c5adaa25182f25e2ddd621a88dd80e
train_20784
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
code
advanced
Task: code Topic: Tool calling, sandboxes, and CI integration Difficulty: advanced Target language: Java Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Java", "developer_needs": [ "evaluation_metrics", "tests_are_truth", "reproducibility", "tooling" ], "moe_experts": [ "data_curation_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
ef257f193ef93ba6817239ad9d878c20be489346
train_20785
2026-01-01T00:00:00
Latency, cost, and reliability optimization
data_pipeline
advanced
Task: data_pipeline Topic: Latency, cost, and reliability optimization Difficulty: advanced Target language: TypeScript Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "evaluation_metrics", "repo_scale_reasoning", "auditability", "documentation" ], "moe_experts": [ "data_curation_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
178bd59d7b1bd4780ef6bfd094279dd8e43833a4
train_20786
2026-01-01T00:00:00
Extended context and repo-scale understanding
design
intermediate
Task: design Topic: Extended context and repo-scale understanding Difficulty: intermediate Target language: Bash Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "Bash", "developer_needs": [ "auditability", "ci_integration", "governance", "tooling" ], "moe_experts": [ "performance_expert", "evaluation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4708e0b2bc32a49c3c79170cb4078cc629a9b786
train_20787
2026-01-01T00:00:00
Multimodal dev workflows (docs, diagrams, traces)
review
expert
Task: review Topic: Multimodal dev workflows (docs, diagrams, traces) Difficulty: expert Target language: Rust Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Rust", "developer_needs": [ "tests_are_truth", "auditability", "repo_scale_reasoning", "cost_latency_tradeoffs" ], "moe_experts": [ "evaluation_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
7e5d55699228e279ea75494353e9daba51a66a50
train_20788
2026-01-01T00:00:00
SWE-bench style real-repo evaluation
failure_analysis
intermediate
Task: failure_analysis Topic: SWE-bench style real-repo evaluation Difficulty: intermediate Target language: TypeScript Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Failure: - Initial patch broke edge case Reflection: - Missing zero-input guard Correction: - Add explicit validation + test
[]
{ "target_language": "TypeScript", "developer_needs": [ "evaluation_metrics", "reproducibility", "ci_integration", "documentation" ], "moe_experts": [ "data_curation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
bea4c214cf1a535cf8e3bfb63d9638eeb8a485dd
train_20789
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
design
intermediate
Task: design Topic: Model merging, distillation, and continued pretraining Difficulty: intermediate Target language: C# Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[]
{ "target_language": "C#", "developer_needs": [ "cost_latency_tradeoffs", "reproducibility", "tooling", "security_gates" ], "moe_experts": [ "security_expert", "agentic_systems_expert" ], "governance": { "audit_required": true, "tests_required": true } }
5059137b0be752a253b0f767e94ae18c8676ae37
train_20790
2026-01-01T00:00:00
Extended context and repo-scale understanding
patch_diff
advanced
Task: patch_diff Topic: Extended context and repo-scale understanding Difficulty: advanced Target language: C# Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Patch (diff-style): ```diff - if x == 0: - return 1/x + if x == 0: + raise ValueError('division by zero') ``` Acceptance: - Tests pass - No new regressions
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "C#", "developer_needs": [ "repo_scale_reasoning", "evaluation_metrics", "auditability", "ci_integration" ], "moe_experts": [ "data_curation_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
03104706cf80a4e768bdd597814935dcb5e53654
train_20791
2026-01-01T00:00:00
Agentic coding systems (plan→edit→test→reflect)
agent_loop
expert
Task: agent_loop Topic: Agentic coding systems (plan→edit→test→reflect) Difficulty: expert Target language: Python Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Python", "developer_needs": [ "repo_scale_reasoning", "governance", "reproducibility", "documentation" ], "moe_experts": [ "evaluation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
1acf97d6d884c95d20272f4a27c0bb6ae260c162
train_20792
2026-01-01T00:00:00
Governance, provenance, and licensing for code data
review
expert
Task: review Topic: Governance, provenance, and licensing for code data Difficulty: expert Target language: C# Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Review: correctness, security, performance, governance
[]
{ "target_language": "C#", "developer_needs": [ "repo_scale_reasoning", "security_gates", "tooling", "ci_integration" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
5f740b259d0cbf946d86b10f1bd805c985336926
train_20793
2026-01-01T00:00:00
Reasoning-first coding models and tunable deliberation
agent_loop
expert
Task: agent_loop Topic: Reasoning-first coding models and tunable deliberation Difficulty: expert Target language: Bash Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Loop: Plan → Edit → Test → Reflect → Human gate
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "Bash", "developer_needs": [ "documentation", "tests_are_truth", "cost_latency_tradeoffs", "reproducibility" ], "moe_experts": [ "data_curation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
d078c570307305a501746be2d5f05463b43d2aaa
train_20794
2026-01-01T00:00:00
Latency, cost, and reliability optimization
code
advanced
Task: code Topic: Latency, cost, and reliability optimization Difficulty: advanced Target language: Python Context: Large monorepo with flaky tests and strict CI. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Scaffold: ```python def agent_loop(plan, edit, test, issue, max_iters=4): history = [] p = plan(issue) for _ in range(max_iters): patch = edit(issue, p) ok, report = test(patch) history.append({"plan": p, "ok": ok}) if ok: return patch, history p = p + " | refine" return patch, history ```
[]
{ "target_language": "Python", "developer_needs": [ "cost_latency_tradeoffs", "repo_scale_reasoning", "documentation", "reproducibility" ], "moe_experts": [ "data_curation_expert", "governance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
bc7a8454d4bfe4080131ac1a52ea78e9c0bf3f4a
train_20795
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
eval
intermediate
Task: eval Topic: Tool calling, sandboxes, and CI integration Difficulty: intermediate Target language: Rust Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Eval: pass@k, time-to-green, regressions, diff size
[]
{ "target_language": "Rust", "developer_needs": [ "governance", "security_gates", "ci_integration", "tooling" ], "moe_experts": [ "coding_expert", "security_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4235c0a47708143c2dcc01d24e29796e1772caa6
train_20796
2026-01-01T00:00:00
Code-specialized model families and sizing tradeoffs
compare
expert
Task: compare Topic: Code-specialized model families and sizing tradeoffs Difficulty: expert Target language: JavaScript Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "JavaScript", "developer_needs": [ "security_gates", "documentation", "repo_scale_reasoning", "governance" ], "moe_experts": [ "performance_expert", "coding_expert" ], "governance": { "audit_required": true, "tests_required": true } }
81348d8259fcea6fb92100f39f2eaed115d36073
train_20797
2026-01-01T00:00:00
Tool calling, sandboxes, and CI integration
compare
intermediate
Task: compare Topic: Tool calling, sandboxes, and CI integration Difficulty: intermediate Target language: Java Context: Research team validating claims against real repos. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Compare: capability, cost, latency, reliability
[]
{ "target_language": "Java", "developer_needs": [ "cost_latency_tradeoffs", "tooling", "ci_integration", "security_gates" ], "moe_experts": [ "evaluation_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
b50574c26689ca0e02bc26cf2b5f94af3cdd3c60
train_20798
2026-01-01T00:00:00
Extended context and repo-scale understanding
data_pipeline
advanced
Task: data_pipeline Topic: Extended context and repo-scale understanding Difficulty: advanced Target language: C# Context: High-traffic service with latency SLOs. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Pipeline: Ingest → Normalize → Filter → Dedupe → Quality → Mix → Audit
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "C#", "developer_needs": [ "reproducibility", "ci_integration", "tooling", "documentation" ], "moe_experts": [ "governance_expert", "data_curation_expert" ], "governance": { "audit_required": true, "tests_required": true } }
2c95a113abfa8ea15d2f6365f17334520e684201
train_20799
2026-01-01T00:00:00
Model merging, distillation, and continued pretraining
design
advanced
Task: design Topic: Model merging, distillation, and continued pretraining Difficulty: advanced Target language: TypeScript Context: Offline/local deployment with limited compute. Produce expert-level, production-ready artifacts.
Facts: - Modern AI coding prioritizes correctness, evaluation, and governance. - Agentic loops with test gates outperform single-pass generation. Design with risks, metrics, acceptance criteria
[ "Identify failing behavior via tests", "Minimize change surface", "Apply patch", "Run targeted + full suite", "Verify no regressions" ]
{ "target_language": "TypeScript", "developer_needs": [ "tooling", "repo_scale_reasoning", "auditability", "reproducibility" ], "moe_experts": [ "data_curation_expert", "performance_expert" ], "governance": { "audit_required": true, "tests_required": true } }
4444e8356dbb20c48683458e92d3143465c889fd