Selected work · experiments · professional projects
Things I’ve built
to learn by doing.
A growing collection of software products, data systems, machine-learning experiments, and machine-vision work.
- 01
Grab
Systems engineering
A real-time cryptocurrency arbitrage engine that turns live exchange rates into a graph and hunts for profitable cycles.
C++20Boost.AsioOpenSSL - 02
Nido
Full-stack product
A self-hosted property-tracking workspace built around the idea that losing data is worse than being briefly unavailable.
GoReact 19TypeScript - 03
WCS
IoT platform
An end-to-end sensor and weather platform, from the firmware on the microcontroller to the charts in the browser.
DjangoNext.jsTimescaleDB - 04
Fantasy RL
Reinforcement learning
A custom reinforcement-learning environment where agents learn to run a fantasy football team on a fixed budget.
PythonGymnasiumMLflow - 05
GLUP24
Deep learning research
Deep-learning models that predict blood glucose levels for people with type 1 diabetes from continuous monitoring data.
PythonPyTorchOptuna - 06
Fantasy API
Data infrastructure
A local-first Python data layer that turns a volatile external feed into a dataset you can build on.
PythonData pipelinesREST