Complete AI Lab Training Pipeline Implementation for Developers

Implement a fully functional AI Lab training pipeline, integrated end-to-end into the existing project, ensuring no skipped, stubbed, or partial delivery of any component.
This comprehensive task requires the agent to perform all steps immediately, including:
– File creation
– Testing
– Report generation
– Opening a PR with full metadata and detailed deliverables
The agent must strictly adhere to the enforcement rules, input/output signatures, and conventions outlined below without exception.
# Primary Requirements
1. Start with non-negotiable backup:
– Create and verify a timestamped backup of the repo root.
– Abort and fail if missing or checksum mismatch.
2. Branching:
– Create a timestamped feature branch for all changes.
3. Backend Implementation:
– SQL migration script to create AI Lab tables supporting SQLite and Postgres.
– TypeScript DB connection and migration utilities with `DB_ENGINE` environment toggle.
– Typed services for experiments, datasets, training runs, sources, and models with specified asynchronous functions.
– Express routes implementing AI Lab API with Zod validation, error wrapping, and consistent JSON responses.
4. Worker Implementation:
– TypeScript worker that polls for pending training runs, claims, and runs them with a deterministic toy trainer for CI.
– Supports cancellation checks, artifact saving with metadata, status updates, and error handling.
– Expose callable functions for testing.
5. Frontend Implementation:
– Sidebar component with nested AI Lab menu and full RTL and keyboard support.
– AI Lab pages and utility functions with unit tests.
– React hook interface for training features tied to API.
– API services abstracted with base URL from environment variables.
6. Tests:
– Unit and integration tests for every new or modified function or file with coverage thresholds >= 90%.
– Integration tests of all AI Lab endpoints and frontend components.
– Use Jest/Vitest and Supertest suitably.
7. CI Pipelines and Verification Script:
– GitHub workflows updated or added with ordered jobs for lint, type-check, unit test, integration, build, and smoke.
– Smoke job runs backend and worker, verifies artifact upload, and executes the verification shell script.
– `verify_delivery.sh` must run all checks, start services, conduct smoke test, and exit 0 only on success.
– Generate machine-readable JSON and Markdown reports.
8. PR Requirements:
– Include backup report with exact filename and checksum.
– Include `ci_fix_summary.md` representing before/after diffs.
– Attach `BOT_CODER_DELIVERY_REPORT.md`, `PROJECT_FINAL_VERIFICATION_REPORT.md`, logs, `backup.txt`, and coverage reports.
– PR description must start with the exact delivery verification header including backup, checksum, commit hash, and branch.
– At least one human review and approval required; no merging without it.
9. Failure Handling:
– Any acceptance criteria failure must create a blocking issue with failing logs and tests; no merge allowed.
# Required Files, Functions, and API Specifications
– Implement precisely the files and function signatures specified in section 2, including exact filenames and export signatures for backend and frontend.
– All backend services, routes, and utilities must handle errors properly and validate input with Zod.
– The worker must support running jobs deterministically and updating their status atomically.
– Frontend components and hooks must support navigation, state management, and API communication as outlined.
# Acceptance Criteria (must all be true before marking done)
– Verified backups
– Properly named branches and PR headers
– Complete, passing lint/type-check/tests with coverage
– Functional endpoints and UI
– Successful smoke tests
– Artifact saved on disk and DB
– Proper report generation
# Required Local Commands
Run all listed commands to create backups, run lint/type-check/tests, apply migrations, start backend/worker, then run verification script, saving outputs in `reports/logs/`.
# Enforcement Rules
– No stubs or placeholders.
– All code must be modular functions with proper types and documentation.
– Only non-destructive changes allowed.
– Verification steps and reporting rigorously enforced.
# Final Message on Success
Print exactly the provided success message with backup path and artifact location.
Paste and run the above instructions fully now, produce the PR with all deliverables, and wait for required human approval. If any checks fail, open a blocking issue with logs and test results.
Upon completion, share the PR URL and the JSON output from `PROJECT_FINAL_VERIFICATION_REPORT.json` for immediate evaluation and targeted fixes.
# Output Format
Provide all necessary files, code, tests, and documents as specified in the deliverables section.
Include:
– Complete implemented functions and migrations.
– Tests (unit and integration).
– React components, pages, hooks with tests.
– CI workflow definition.
– Verification shell script and optional TypeScript script.
– Reports and logs.
– Documentation markdown files with usage.
Attach a detailed PR description starting with the exact delivery header.
Print the success message only if all acceptance criteria are met.
# Notes
This prompt is an uncompromising enforcement prompt expecting full implementation. Do not skip or partially implement any part.
Follow all instructions exactly. Failure to meet any acceptance criteria is treated as a blocking failure.
Execute immediately and fully with no clarifying questions.
# Tags
“AI Lab”, “Training Pipeline”, “Full CI”, “TypeScript”, “React”, “Express”, “SQLite and Postgres”, “Strict Enforcement”
# Name
AI Lab Training Pipeline Completion
# Icon
RocketLaunchIcon
# Category
programming