Leandro Damasio. Welcome to My Personal Page

I am an AI Engineer at Enforce (BTG Pactual Group), based in São Paulo, Brazil.

I build production-grade AI systems for financial and legal domains, where reliability, observability, and governance are not optional, they are foundation. My work focuses on making AI systems safe to operate in real decision-making workflows, especially in regulated environments.

My work sits at the intersection of AI, software architecture, and infrastructure. I design agentic architectures, define runtime control loops, and build evaluation and monitoring pipelines for LLM-based systems. I work hands-on with vector search using pgvector and ParadeDB, prompt and context governance, and secure integration with internal and external data sources.

I place strong emphasis on versioning, auditability, and traceability, ensuring AI systems meet technical, legal, and risk requirements as they evolve in production.

Beyond my role at Enforce, I am active in AI-first development and agentic systems beyond the corporate setting. I am the creator and maintainer of an open-source monorepo at RBX Systems, where I work on AI agents, code agents, and system-level tooling, always with a focus on architecture and long-term maintainability.

I enjoy collaborating across Software Engineering, DevOps, UX, and Product, and I am especially interested in agentic runtimes, memory and context management, and the boundary between model capabilities and system-level control.