Meta × PyTorch OpenEnv Hackathon
Reinforcement-learning agent for energy demand shaping. Global top-800 finish.
From signal to system.
Code as instrument, intent as compass.
Reinforcement learning enthusiast. Currently building intelligent systems at the intersection of code and consequence.
Computer Engineering student at Dhole Patil College of Engineering (2025–2029). I build reinforcement-learning systems and full-stack interfaces, sometimes in the same week, sometimes in the same project.
My work sits where rigor meets craft: research-grade RL environments, ergonomic web tooling, and the unglamorous wiring that makes either feel inevitable.
Based in Pune, India. Freelancing since January 2024.
Five disciplines, one operator. Click a tab to inspect.
Dhole Patil College of Engineering, Pune.
Independent. End-to-end client web apps: design, frontend, deployment.
Reinforcement-learning agent for energy demand shaping. Global top-800 finish.
NeuroBoard: AI smart canvas for handwritten math and diagram refinement.
Smart academic workflow shipped in 36 hours as the lone first-year team in the bracket.