AI integrated into the product
Agents, GraphRAG, and tools grounded in the business context. Evaluations, guardrails, and observability help me understand their behavior and identify where they need to improve.
About
LLM agents, built into real products
I earned my degree in Information Systems Engineering from Universidad Nacional del Sur. Since 2024, I’ve combined independent projects with engineering roles in enterprise software, fintech, and blockchain integrations.
I built systems for merchants and enterprise back offices, connecting services and incorporating fintech and blockchain integrations. Filomena, my final-year project, marked the completion of my degree and grew into an exam platform used by five institutions in Argentina. We developed it as a three-person team.
I lead engineering at Rampy and work closely with the three founders and with specialists in blockchain and integrations. Rampy is an early-stage fintech and blockchain startup whose non-custodial app opens the world of digital assets (swaps, yield vaults, perps and market news) to everyday people through a generative interface that turns complex concepts into simple ones. I own the AI assistant (the agents, their evals and the backend behind them) and the web and mobile app where people use it. Priorities move fast, so I balance getting useful changes into users’ hands with keeping the product easy to maintain and grow.
I treat reliability as something to measure: typed contracts between the agent and the UI, automated tests, eval runs before each release, tracing in production and tools that only run inside explicit limits.
When a small team needs it, I also handle deployment: Coolify, NGINX, Traefik and Caddy.
I like the point where a vague product question becomes a measurable one: what the agent should do, how we test it and what it costs. That is usually where I help a small team most.
What I bring
Connecting AI, product, and systems
Web and mobile experiences
Clear workflows, design systems, and reusable components. I connect interfaces with services and pay attention to perceived speed, accessibility, and error states.
Enterprise software and fintech
Service, permission, and data integrations with explicit contracts. I also connect products with DeFi protocols, markets, and wallets, accounting for the specifics of each operation.
How I work
Engineering standards centered on the user
These standards guide how I turn a need into a working feature, from the first questions to what happens after release.
Start with what the user needs
I first understand what the user needs to accomplish, what constrains the solution, and how we’ll know it works. That guides a clear, accessible, and useful experience.
Build for change
I organize code by domain, separate responsibilities, and define clear contracts. Reusable components and documented decisions make new features easier to understand, test, and maintain.
Improve performance with evidence
I measure the full journey, find where time is lost, and compare the results after each change. I prioritize what users notice: quick startup, responsive interactions, and smooth transitions.
Give failures a way forward
I plan for invalid inputs, permission issues, and failures in external services. I define clear responses, safe choices, and recovery paths. Observability helps me detect problems and act before they escalate.
Verify quality through production
I combine automated tests, integration checks, and human review of the user experience. After release, I monitor real behavior. For AI, I also evaluate responses, tool use, and guardrails against representative scenarios.
My next step
AI and software engineering, with a product focus
I’m looking for fully remote AI Engineer roles building LLM agents, RAG and GraphRAG systems, and the evals that keep them reliable. My core stack is Python/FastAPI, LangGraph, Claude, LangSmith and vector retrieval.
I’m also open to software, backend, and full-stack engineering roles where I can contribute my experience with enterprise systems, web and mobile applications, and integrations. I enjoy contributing to product decisions and seeing features through from initial requirements to production.
I’m open to new opportunities and available for interviews.