Work
Projects I've built, teams I've led, and problems I've solved.
De-Risking Autonomous AI Game Generation Before Betting the Company
Validated autonomous AI game generation feasibility and delivered a working PoC as sole architect
Sole Architect & Implementer
Removing Hidden Architectural Debt Before Scale Broke the System
Removed architectural debt and reduced API latency 40–60% while maintaining feature velocity for a 5-person team
CTO / Technical Lead
Surviving Two Platform Pivots Without Burning the Runway
Navigated App → Web → App pivots and shipped iOS + Android in 4 weeks by reusing ~90% of the web stack
CTO / Technical Lead
Recovering From a Stack Decision That Cost 6 Months
How a mobile-first stack choice created hard performance and SEO ceilings and how a web-native rewrite restored speed and leverage
CTO / Technical Lead
Stabilizing a Live System Under Growth and Fundraising Pressure
Re-architected a fragile production backend while supporting 5× user growth (cut response times by ~80–90% and eliminated single points of failure)
Sole backend owner (technical co-founder)
Audio-Visual Speech Recognition Using OpenCV
Implemented multimodal speech recognition combining audio and visual lip reading for robust recognition in noisy environments
ML Engineer (Research)
Diagnosing and Rebuilding a Computer Vision Pipeline Under Real-World Constraints
Redesigned age/gender estimation pipeline after systematic failure analysis, achieving 82% gender accuracy on degraded CCTV footage
ML Engineer / System Designer
Building Production-Ready Gesture Recognition Under Real-Time Constraints
Designed and validated real-time dynamic gesture recognition system with 90% accuracy and MediaPipe integration
ML Engineer / Researcher
Scaling Edge ML Infrastructure Through Automated Quantization
Built automated quantization-aware training pipeline reducing model deployment time from days to hours
AI Intern at EdgeNeural
Multi-Agent Decision-Making via Reinforcement Learning & Game Theory
Mentored research on emergent cooperation and altruism using RL agents in game-theoretic environments
Research Mentor, Brain & Cognitive Society, IIT Kanpur
Implementing SRGAN for Photo-Realistic Image Super-Resolution
Built TensorFlow implementation of SRGAN combining perceptual and adversarial losses for realistic image upscaling
ML Engineer (Research)
Implementing WaveGAN for Raw Audio Waveform Generation
Built TensorFlow implementation of WaveGAN for end-to-end audio synthesis from random latent vectors
ML Engineer (Research)
Disentangling Math vs Story Reasoning in the Human Brain
Used fMRI and machine learning to identify neural basis for mathematical vs narrative reasoning with 97% accuracy
Researcher, Neuromatch Academy (Computational Neuroscience)