Add AI Product Manager archetype

[SaharPak]

Why

The README advertises 6 archetypes; only 3 are implemented. AI Product Manager is the second-most-requested missing archetype. It's distinct from the existing product-manager archetype in that the keyword and tooling weight should differ — AI PMs need:

• AI/LLM-specific keywords (RAG, agents, prompt engineering, evals, model serving, fine-tuning, MCP, vector DB)
• Stronger publicProof weight (AI PMs often have side projects, GitHub repos, content)
• Stronger shippedEvidence weight (AI features are notoriously hard to ship — proof of one matters more)

The current product-manager archetype is tuned for general PM work and under-weights AI evidence.

What to do

1. Copy packages/intelligence/src/archetypes/product-manager.ts as a starting point.
2. Augment the keyword set with AI-specific terms (sample list below).
3. Adjust evaluationWeights to emphasize shippedEvidence and publicProof.
4. Add tests.
5. Submit a PR.

Suggested starting keywords (add to PM's existing 18)

• RAG, LLM, agents, agentic, prompt engineering, evals, fine-tuning
• MCP, Claude, GPT, Cursor, Claude Code
• Vector database, embeddings, semantic search
• AI infrastructure, model serving, inference
• Evals, evaluation harness, A/B testing for AI

Effort

M (2–4 hours).

Impact

High. The user population for this archetype overlaps heavily with the project founder's own profile. Landing this archetype and dogfooding it against the founder's own CV is a useful end-to-end test of the full pipeline.

[IKetutWidiyane]

Hi! I'd like to work on this issue. Could you please assign it to me? Thanks!