i am freshman at college studying computer science.
currently i am working as a
- research intern at the Language Technology Research Centre, IIIT H (on edge ai & inference)
- sde intern at a stealth startup (on ai agents)
previously i had worked as an intern elsewhere to improve a moderation risk engine used by BuyMeACoffee(YC W19).
- resume / github / twitter
1. spheni: vector search library with product quantization and residual encoding
2. rayfloat: a multithreaded, bvh accelerated cpu path tracer
3. sarvam.c: an inference runtime for sarvam-1 model, with support for sentence-piece tokenizer
1. sameshi: a ~1200 elo chess engine that fits within 2kb of c
2. dunefetch: if neofetch and a falling sand simulator had a baby
3. lookin: a dead simple linux syscall tracer
4. frogemm: a small gemm library in c to do fast matmul
5. inwhale: a quantization library made from scratch
1. grudgepixel: a pixel art editor with animation support (made it when i was 14, with c/raylib)
2. amine: distraction blocker but it actually takes control of your hardware
3. yars: reddit scraper without api keys and all
4. touchgrass: a chess engine playground in python
5. froginfer: an inference runtime for llama2 models in c
1. [dec 2025] why my ray tracer got slower before it got faster
2. [nov 2025] exploring garbage collector and how it works
1. [feb 2026] why ai doesnt make me anxious
2. [oct 2025] impermanence