Backend & AI engineer building reliable LLM systems.

Fifteen years across backend systems, EU digital identity, and embedded hardware, now focused on AI engineering. I care about the part most demos skip: making the system around the model dependable, and keeping it from getting exploited.

Experience
Backend & AI engineering, on contract Adversarial testing of production LLM systems, with findings mapped to the regulation they answer to. Before that, EU digital product passports and verifiable-credential wallets that passed EBSI Conformance Testing in full — and the access-control layer and audit trail of a multi-tenant platform, where letting the wrong tenant through once is a breach rather than a bug. 2022 —
Research engineering, university institute Three multi-sensor systems built end to end — industrial, human, marine — each one sensor hardware through firmware to the server and the interface researchers actually use. The environmental network I built around a working port still runs, and I still keep it running. 2011 —
iOS, on contract Sole engineer on a production market-tracking app for four years. Inherited an unstable MVP and rewrote it; state stayed the client's, computed on-device against a live feed and held together across reconnects, backgrounding and going fully offline. 2018 – 2022
Writing
Assume the guardrail fails The best-scoring open guard model answers 91% correctly on public benchmark prompts and 33.8% on ones it has not seen. That gap should change how you build, not which model you pick. 2026 AWS Blocks and the hard 20% A framework that makes local RAG feel like magic, until you deploy. Where the abstraction held, where it leaked, and what I'd reach for it for. 2026 Making an AI agent reliable Three times building one agent, the same bug bit me. Each time the fix was the same principle: never let the model hold a value that has to be exact. 2026
Selected work