I work on backend and data systems: the kind with real users, real traffic, and real consequences when something is wrong.
That’s been about a decade now, mostly helping companies build cloud platforms and get AI systems into production. Agentic platforms, RAG applications, enterprise workflow automation, and the unglamorous plumbing that decides whether any of it survives contact with actual users. I’ve done this at startups, at consultancies, and inside large corporations, which is a useful way to learn that the hard problems are rarely the technical ones.
I also lecture on a Master’s programme in Data Science and AI. Teaching is an excellent way to find out how much of what you believe you can actually justify.
What I write about here
Mostly applied AI and backend engineering: what works in production, what quietly doesn’t, and where the gap between the demo and the deployment turns out to be. When a post makes a measurable claim, I try to run the thing and publish the code so you can check my work rather than take my word for it.
Sometimes I write about none of that. It’s my blog, so occasionally it’s a reflection, an argument with something I read, or a subject with no obvious connection to my day job. I’d rather that than pad the archive with posts I didn’t want to write.
Working together
At Iwana Labs we build and operate production AI systems for clients. It’s a small consultancy in Madrid, which means the people you talk to are the people who do the work. No handoff to a delivery team you never met.
If something here maps onto a problem you’re trying to ship, that’s a good reason to get in touch. The fastest routes are LinkedIn or the contact form at iwanalabs.com.