Showing results for "regression-suite"
Create a safety regression suite for tool-using agents: prompt injection tests, permission misuse tests, and data leakage tests. Gate deployments on this suite.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Build a dependency graph indexer and an impact analyzer that estimates blast radius of a change and picks appropriate test suites and reviewers.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Design golden tests for key outputs and an agent that compares before/after behavior. Include tolerance rules and how to prevent “golden drift” over time.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Build a refactoring agent that only changes code when it can show measurable gains (latency, memory, error rate). Include baseline capture and regression detection.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Create a test selection algorithm that runs only relevant tests for a diff but guarantees critical coverage (safety suite always runs). Include calibration and drift checks.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Design an agent that expands benchmarks as new features land: add workloads, track performance trends, and alert on regressions. Include benchmark governance.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Create an agent that turns bug reports into minimal reproducible cases, adds regression tests, and proposes fixes. Include a triage rubric and severity mapping.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Design characterization tests for legacy code before refactoring: capture I/O behavior, edge cases, and performance baselines. Use them as a gate for changes.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Design a loop to detect memory leaks: reproduce with stress tests, capture heap snapshots, propose fix, and verify stability. Include regression tests.
Author: Assistant
Category: safe-self-improving-ai | Model: gpt-5.2
Design a sanitization pipeline for retrieved content: strip instructions, isolate quotes, and prevent tool-use hijacks. Include adversarial test cases and regression suite.
Author: Assistant
Category: research-bot | Model: GPT-5.2
Design an evaluation strategy that avoids overfitting: holdouts, rotating test sets, adversarial sets, and blind evaluation. Include rules for when to refresh benchmarks.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
Design defenses against prompt injection for tool-using agents: content provenance, allowlists, tool policy, and sandboxing. Include a suite of adversarial prompts for regression testing.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
Create a post-mortem template tailored to AI regressions: data/prompt/model diffs, evaluation gaps, monitoring misses, and remediation tasks. Include a ‘lessons to tests’ section.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
Design a hallucination reduction plan: retrieval augmentation, answer verification steps, consistency checks, and refusal behaviors. Include evaluation metrics and regression tests.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
Create a safety regression suite: prompt injection tests, data leakage tests, refusal/guardrail tests, and policy adherence checks. Include how to maintain and evolve the suite over time.
Author: Assistant
Category: recursive-ai-safety | Model: GPT-5.2
Create a defense plan against prompt injection when agents consume untrusted text: content provenance, instruction isolation, and safe tool policies. Provide a red-team test suite.
Author: Assistant
Category: agent-architecture | Model: GPT-5.2
Design layered guardrails: policy rules, tool allowlists, output filters, and human escalation. Include regression tests to prevent guardrail drift during iterations.
Author: Assistant
Category: agent-architecture | Model: GPT-5.2
Create a benchmark suite that measures planning quality, tool-call correctness, and end-to-end success. Include scoring rubrics, difficulty tiers, and anti-overfitting practices.
Author: Assistant
Category: agent-architecture | Model: GPT-5.2
Design a CI pipeline for hardware: lint rules, CDC checks, reset checks, basic formal proofs, and regression simulation tiers. Include pass/fail gates and artifact retention for debug.
Author: Assistant
Category: fpga-asic-design | Model: gpt-4o
Create operational runbooks: common failures, triage steps, rollback, and user comms. Include SLO breaches, tool outages, and prompt regressions.
Author: Assistant
Category: agent-architecture | Model: GPT-5.2
Create an eval-first LLMOps design: golden sets, adversarial tests, continuous regression, cost/latency tracking, and release gates. Include a ‘model change control’ policy.
Author: Assistant
Category: ai-strategy-2026 | Model: gpt-4o
Create a quantization evaluation suite (GPTQ/AWQ/RTN): perplexity, zero-shot accuracy, calibration set selection, and layer-wise sensitivity. Output deployment guidelines by architecture and hardware ...
Author: Assistant
Category: model-compression-LLM | Model: gpt-4o
Given a product brief, have ChatGPT write acceptance criteria and test outlines, ask Cursor to generate unit/integration tests and the minimal code, then let Antigravity agents execute the suite and o...
Author: Assistant
Category: quality-engineering | Model: gpt-4o
ChatGPT drafts prompts and adversarial tests; Cursor integrates an eval harness; Antigravity schedules nightly evals and posts regressions with diffs. Include versioning and approval flow.
Author: Assistant
Category: mlops-llm-quality | Model: gpt-4o
Define performance budgets and benchmarks; ChatGPT proposes profiling plan; Cursor instruments code (e.g., pprof/Flamegraphs); Antigravity runs load tests, flags regressions, and files targeted PRs. O...
Author: Assistant
Category: performance-engineering | Model: gpt-4o
Build an eval that scores ground-truth attribution (exact passage match), answer faithfulness, and coverage. Provide dataset schema and a nightly regression plan.
Author: Assistant
Category: evaluation-frameworks | Model: gpt-4o
Act as a meta-analyst. Provide a DerSimonian–Laird vs REML plan, heterogeneity measures, subgroup/meta-regression, and publication bias checks (funnel/Egger/trim-and-fill).
Author: Assistant
Category: biostats-methods | Model: gpt-5
Define a push-button signoff pipeline: reproducible EDA containers, golden rule decks, regression checklists, run orchestration, artifact retention, and ECO automation for timing/IR/DRC. Output a CI/C...
Author: Assistant
Category: chip-design | Model: gpt-4
Provide a timing ECO guide: buffer sizing rules, cell Vt swaps, re-route constraints, shielding for aggressors, hold-fix ordering, and checks to avoid IR/EM regression. Include a before/after metrics ...
Author: Assistant
Category: chip-design | Model: gpt-4
Define a benchmarking suite for variational algorithms: datasets, seed control, metrics (energy error, convergence time), and reproducibility notes. Output: CSV spec (metrics) and YAML (suite config).
Author: Curioforce Corp. Corp.
Category: Quantum Tech | Model: gpt-5-thinking
You are a robotics CI lead. Design a sim-in-the-loop pipeline for perception + planning stacks. Deliver: scenario library spec (lighting, occlusion, rare edge cases), metrics (success %, collision=0, ...
Author: Tsubasa Kato
Category: Engineering | Model: GPT-5 Thinking
Plan a simulation for a safe pick-and-place arm in a workshop. Deliver: kinematics, obstacles, planner choice (RRT*/PRM), KPIs (cycle time, collisions=0), digital twin setup, HMI cues. Output: scene f...
Author: Tsubasa Kato
Category: Simulation | Model: GPT-5 Thinking
Create a science fiction adventure set on Mars centered around two teenage protagonists, Kael and Lyra, who uncover a hidden signal beneath Olympus Mons. The story explores their emotional journey, th...
Author: [email protected]
Category: story,story plot,Mars,space,science fiction | Model: GPT-4o,o3-mini,o1