Francisco Ferreira

Python, SQL, AWS, LLMs, RAG, Prompt Engineering, LangGraph, Next.js, Supabase. Mostly backend/AI but ship full-stack when needed.

LangGraph for stateful agents, advanced RAG evaluation patterns, and how to make LLM outputs reliable enough to stop babysitting them in production.

PromptEval — a production-grade prompt evaluator. Scores prompts 0-100, finds root causes, and proposes surgical fixes based on expected vs observed behavior.

Discussing production LLM failures, prompt debugging, RAG architecture. Also open to feedback on tools I'm building for AI developers.