Daniel Enache

Meet Daniel

1 following

Most AI systems die between the demo and production. I build the ones that survive.

I ship production AI, not demos — and everything I claim is public and verifiable:

→ agentic-rag-mcp (open source) — a multi-agent RAG server over the Model Context Protocol. Every answer cites its exact source chunk or the agent refuses to answer. The eval suite — groundedness, citations, refusal correctness — runs in CI, so a regression fails the build before it reaches a user.

→ mcp-vitals (open source) — a CLI that grades any MCP server A–F for reliability and agent-usability. I pointed it at the official reference servers: two earned an A, one got an F. Live report: enached134-ctrl.github.io/mcp-vitals

→ drumvia.com — a live SaaS marketplace I designed, built and operate solo, end to end: auth, Stripe payments, KYC, real-time bidding, row-level security.

For clients: RAG with real citations, MCP servers, LangGraph agents, and the eval harnesses that prove an AI feature works before your users test it for you. Also 2+ years operating AI content-generation pipelines.

Tell me what your AI is supposed to do — I'll tell you how to prove it does it.