I’m Joaquin Casanova. I build data platforms and AI systems that work in production.
My background is in physics, which means I learned early to reason from first principles before touching a computer. That habit has been more useful than any framework — it’s how I approach data architecture, model design, and the kind of AI integration work that breaks when the assumptions don’t hold.
I’ve spent the last several years building at the intersection of data engineering, MLOps, and AI systems, primarily on AWS. I’ve designed batch inference pipelines, built internal data platforms, led teams through the transition from manual data processes to automated ML workflows, and more recently started building with MCP and agent architectures to connect AI to real internal systems.
My current focus: production AI systems, internal intelligence layers, and the data infrastructure that makes AI usable — not just demonstrable.
I write about what I build. The posts here are technical and specific.
If you’re building something in this space and want to work together, reach out or visit the Work With Me page.