Aleph One.

The work

Senior engineer across three production products — Beana AI (AI search platform), CritiqueMatch (consumer product with paid services), and an internal Calls App — owning features end-to-end from backend architecture to UI, and working directly with the client to turn feedback into scoped, shipped features. On Beana AI I drove the AI direction: retrieval and agent architecture, model selection, and the eval strategy that kept LLM quality measurable.

Key impact

• —Cut AI search latency from 20+ seconds to ~3–4 seconds: profiled the retrieval and ranking path and fixed the bottlenecks (to 6–7s), then parallelized tool calling to get under 4 seconds.
• —Raised RAG pipeline accuracy from ~30% to 90%+ on QA test suites validating filters and LLM outputs.
• —Grew automated test coverage from zero to ~75% across both frontend and backend.
• —AI search running in production over a catalog of 100k+ items, serving real users daily.

What I did

• —Shipped features across three production products — Beana AI, CritiqueMatch, and an internal Calls App — owning them end-to-end from backend architecture to UI.
• —Drove the AI direction of Beana AI, the flagship AI search platform: retrieval and agent architecture decisions.
• —Set up AI development workflows that let even non-developers run agents producing properly structured code — automatically reviewed, and blocked when checks fail.
• —Worked directly with the client to turn feedback and feature requests into scoped, prioritized features — owning delivery end-to-end.

Stack