AGI Jobs v0 (v2) — Demo → AlphaEvolve v0

Explore a bounded local search over job-allocation heuristics. Compare candidate utility, modeled cost, fairness and acceptance against a supplied baseline, then inspect why guardrails accepted or rejected each generation.

Run the implemented experiment

python demo/AlphaEvolve-v0/run_demo.py run --seed 7 --generations 5 --output /tmp/alphaevolve-report.json
python -m pytest demo/AlphaEvolve-v0/tests

The root launcher uses alphaevolve_runner.py and alphaevolve/. It generates synthetic agents/jobs and uses ScriptedLLMClient to propose bounded heuristic changes. The evaluator computes outcomes locally; no live model, wallet, contract transaction or real purchase occurs. Cost and utility are model quantities, not measured provider bills or earned revenue. A passing guardrail authorizes only promotion within this local search experiment.

The JSON report includes baseline/final/champion metrics, per-generation decisions, guardrail reasons and explicit execution-scope fields. Existing alphaevolve_demo/ is a separate implementation; do not assume a test of this launcher qualifies it. The broader integration diagram below describes the design context, not an automatic production route established by this command.

Systems Map

flowchart LR
    Operators((Mission Owners)) --> demo_AlphaEvolve_v0[[Demo → AlphaEvolve v0]]
    demo_AlphaEvolve_v0 --> Core[["AGI Jobs v0 (v2) Core Intelligence"]]
    Core --> Observability[[Unified CI / CD & Observability]]
    Core --> Governance[[Owner Control Plane]]

Working With This Module

1. From the repository root run npm install once to hydrate all workspaces.
2. Inspect the scripts under scripts/ or this module's package.json entry (where applicable) to discover targeted automation for demo/AlphaEvolve-v0.
3. Execute npm test and npm run lint --if-present before pushing to guarantee a fully green AGI Jobs v0 (v2) CI signal.
4. Capture mission telemetry with make operator:green or the module-specific runbooks documented in OperatorRunbook.md.

Directory Guide

Key Directories

• alphaevolve
• alphaevolve_demo
• config
• tests
• web

Key Files

• __init__.py
• alphaevolve_manifest.json
• alphaevolve_runner.py
• alphaevolve_summary.json

Quality & Governance

• Every change must land through a pull request with all required checks green (unit, integration, linting, security scan).
• Reference RUNBOOK.md and OperatorRunbook.md for escalation patterns and owner approvals.
• Keep secrets outside the tree; use the secure parameter stores wired to the AGI Jobs v0 (v2) guardian mesh.

Next Steps

• Review this module's issue board for open automation, data, or research threads.
• Link new deliverables back to the central manifest via npm run release:manifest.
• Publish artefacts (dashboards, mermaid charts, datasets) into reports/ for downstream intelligence alignment.