About Juan José Expósito
Freelance AI Engineer. PhD in Engineering. 22 years in oil & gas. Based in Madrid, Spain.
The trajectory
I spent more than two decades as a Drilling Engineer — working with Repsol, ADNOC, and Moeve across some of the most operationally complex environments in the industry. That's where I learned what it means to work with high-stakes technical documentation, dense procedures, and institutional knowledge that lives in people's heads instead of where it should be.
In 2020 I made a deliberate pivot. I got certified as a Python developer, started building AI systems, and discovered that the problems I'd spent 22 years living with — buried knowledge, slow information retrieval, manual work that shouldn't be manual — were exactly the problems AI could solve.
Today I build those solutions professionally.
What I bring to each project
-
Operational context
I've worked in the environments where your documentation lives. I understand drilling procedures, safety-critical workflows, and the real consequences of information gaps. You don't need to explain your domain to me.
-
Engineering rigor
PhD in Engineering from Universidad Politécnica de Madrid. Certified Python developer. I approach AI projects with the same structured methodology I used in engineering: define the problem, validate constraints, measure results.
-
Production-grade implementation
My technical stack — Python, FastAPI, LangChain, Weaviate, Pinecone, OpenAI, Anthropic, PostgreSQL, n8n, Make — is chosen for systems that need to keep running after handoff, not for demos.
-
Breadth across domains
Beyond AI for industrial documentation, I work on financial automation and blockchain strategy. Three areas where domain expertise makes the difference between a solution that works and one that almost works.
Areas of expertise
AI for Technical Documentation RAG systems, AI assistants, and knowledge management tools for engineering and industrial organizations. The core problem: critical knowledge is buried in documents and nobody can find it fast enough. Explore this area →
AI Automation for Personal Finance Expense classification agents, financial reporting automation, and budget visibility tools. Practical implementations for individuals and small teams. Explore this area →
Blockchain & Tokenization Advisory, technical content, and implementation guidance for teams working with blockchain infrastructure, digital assets, and tokenization strategy. Explore this area →
Selected proof of work
- Built a RAG system for Quiet Links that enabled natural language queries across 200+ academic papers — delivered in 6 weeks, saving an estimated year of solo development
- Developed an AI expense classification agent reducing manual categorization time by 90% — 100+ transactions per month, zero errors
- Mentored developers through the PyBites Python Mentoring Program, guiding them from zero to production RAG systems
- Deep involvement in blockchain and DeFi: author of a complete Vyper smart contract course covering DeFi Vault architecture, formal invariant proofs, and security patterns
What a client said
"Without exaggeration, I learned the intricacies of developing RAG systems in 6 weeks that would have otherwise taken 6 months to 1 year on my own. Juanjo's deep knowledge of Python development, AI applications, and infrastructure were mind blowing."
Tim Gallati — Founder, Quiet Links Library
Let's talk
If you're evaluating whether we're a good fit, the fastest way to find out is a 30-minute call.