Emmanuel Yator

I build products end to end—from database schema to deployed frontend—and I do it fast. My background spans React, TypeScript, FastAPI, Flutter, Python, and Vue.js, but my focus is the hard part: wiring AI into real applications that work in production.

Over the past year, I designed and shipped a full-stack AI platform with multi-LLM orchestration, real-time WebSocket analysis, a RAG knowledge layer, and third-party API integrations. It runs on a FastAPI backend with async SQLAlchemy and Supabase Auth, and it’s the kind of deliberately production-grade work I do even solo.

I’ve put that same ownership into FineChess, my AI chess coaching platform—an end-to-end “Jarvis for chess” that translates Stockfish UCI outputs into structured evaluations and then into coaching language per move. I built a multi-LLM coaching pipeline (Gemini 2.0 Flash primary with Groq/LLaMA fallback), integrated Chess.com and Lichess APIs for game import, and added 6+ AI-powered features including persona-based coaching and voice coaching via Web Speech API.

When I’m not building AI systems, I ship clean product features: at Agile Solutions, I built production Vue.js components in an Agile sprint workflow and delivered into tight release cycles. I also completed an Errands Management System with Django—handling dependency resolution, migrations, and deployment configuration from scratch—so I understand both the code and the reality of getting it live.