Machine learning, robots, and thinking clearly

I'm Stan, an ML Engineer/Data Scientist in London working on LLM systems and agents, lately exploring embodied AI.

I spent the past year at Dojo, building a forecasting pipeline and a pricing-optimisation engine. Prior to that I was at Faculty AI, building LLM systems, and I've written about them for DataCamp: advanced RAG, distillation, and deployment. I hold an MSc in Artificial Intelligence from King's College London, and I spent a summer at Seoul National University building a robot's vision system, a YOLO model that recognised food from a live camera and prompted the robot to react in real time.

Lately I've been drawn to where agents meet robotics: LLM systems and tool-use agents on one side, robot learning and imitation policies on the other. What interests me is the distance between a policy that works in simulation and one that works on real hardware.

Outside of tech, I'm better at tennis than my ranking suggests and an amateur photographer. I read more than I finish, usually about incentives and why things are the way they are.

I mostly write to think out loud and follow my interests wherever they lead. You can see what I've been reading on the reading log, and if you want to follow along, grab the RSS feed.