Location: Porto Alegre, Brazil (GMT-3)

Remote: Yes

Willing to relocate: Yes

Technologies: Python, Ruby on Rails, OpenAI Agents SDK, agentic/multi-agent systems, RAG/Agentic RAG, evals (LLM-as-judge + code-based), Anthropic/OpenAI/Etc APIs, LiteLLM, PostgreSQL + pgvector, Redis, Sidekiq, React/Next.js, TypeScript, Node.js, AWS, Docker

Résumé/CV: https://gerson.is

Email: gersonazgo@gmail.com

AI Engineer and two-time founder, ~20 years in software. My whole career has been about being a sole engineer or leading small teams (4-5 people), gathering requirements from non-technical people, and turning those into production applications.

Currently I'm a sole AI engineer working directly with clinicians at a startup building a clinical "between sessions" AI agent on WhatsApp. I built it as a multi-agent system: a router assesses each conversation for clinical risk and directs it to specialized agents, with strategies to escalate high-risk cases to human professionals. I also built the RAG pipeline and an evaluation framework from scratch (LLM-as-a-judge plus code-based evaluators, synthetic trace generation, continuous production-trace analysis, all with Arize Phoenix), drawing on clinical safety benchmarks like VERA-MH (Spring Health) and TherapyGym.

My background pairs engineering with business (B.A. in Marketing), so the customer-facing half is something I'm very comfortable with.