Build it to understand it
I kept reading “the clearing house novates the trade” and realising I didn’t understand it. So I built one. The hard path is where the complexity actually lives.
I build the unglamorous infrastructure — clearing engines, virtualisation tooling, data platforms — the layer people only notice when it breaks. Mostly Rust, Python and C. Almost always in public.
I kept reading “the clearing house novates the trade” and realising I didn’t understand it. So I built one. The hard path is where the complexity actually lives.
Rust makes me slower and surer. I’d rather ship tabular Q‑learning that works than deep RL that hand-waves. Scope is a design decision, not a compromise.
Most of this is open source and a fair bit is unfinished. Building in public means letting people see the seams — including the archived experiments.
Lovely Professional University, Punjab. Learning the structured way — DSA, ML — while the projects teach me the rest.
Member, contributor and maintainer. LFX funded my travel to the Open Source Event 2026 in Mumbai, June 16–18.
Point-in-time data platforms, a reinforcement-learning paper trader, and a clearing engine that refuses to drift from its golden file.