About — field notes
Hariom Lohar
Self-taught from India, rebuilding intelligence from scratch — Harvard CS50P 2026, 8h a day, committed in public.
Field notes
Who does the work
Self-taught from India, rebuilding intelligence from scratch — one proof, one commit at a time.
I finished 12th grade and chose the longer road — no bootcamp, no shortcut. The day goes into first principles: mathematics from axioms, derivatives by hand, then neural networks built the same way — a Value graph, backprop and gradient descent written in NumPy before touching a framework.
micrograd_hk is that work as a package you can pip install. Now: CNNs, Transformers and PyTorch. All in public, every day.
- Location
- India · IST
- Focus
- AGI from first principles
- Log
- Day — of —
- Open to
- Freelance & collabs
Three rules
No shortcutsDerive before you import
Do not use what you cannot derive. Math from axioms first, then architectures by hand in NumPy.
Eight hours, every day
One deep-work block, documented in AGI_Research so the log stays honest. Committed daily.
Build in public
Errors, proofs and small wins are all commits. The notebook stays open on GitHub.
Let’s build something that lasts.
Backends that stay up, apps that feel native, and search over your data — or join SIGMOID and build in public, one log at a time.