Open to opportunities

João Borges

@jooborges

Message

I build production LLM and multi-agent systems for autonomous real-money decisions.

Portugal

Message

What I'm looking for

I'm looking for remote EU or Portugal-based roles where I can build and operate production LLM and multi-agent systems end-to-end, especially systems requiring rigorous evaluation, observability, safety controls, and real-world reliability.

I've built and operated a live Claude-based multi-agent decision layer for real-money prediction-market trading, with 20,600+ logged decisions, outcome-graded evaluation, canary lanes, and fail-closed safety controls. I own the full system in Python and Rust, from orchestration and observability to execution, data infrastructure, and deployment.

At Firmitas, I deliver client websites through multi-agent research, code generation, and adversarial QA pipelines, with 15+ builds shipped through automated quality, security, SEO, and accessibility gates. I've also supported Portuguese-language growth at Bitunix and evaluated search and AI-system outputs for Google through a contract quality project.

Experience

Work history, roles, and key accomplishments

Current

AI Engineer, Autonomous Trading Systems

Current

Independent

Jan 2025 - Present (1 year 7 months)

Designed and operated a live multi-agent LLM decision layer for autonomous trading on prediction markets and crypto, with 20,600+ gating decisions logged over a 45-day production run. Built resilient agent swarms with orchestrator-worker topology, quorum voting, and canary deployment, and developed a Rust-based execution stack for on-chain order signing and real-time data processing.