Ankit Sengupta

I'm a CS sophomore at VIT Amaravati specialising in AI & ML, with hands-on experience designing distributed systems, autonomous multi-agent pipelines, and high-performance AI-integrated applications. I don't just study machine learning — I ship it. My work spans the full stack: C++ firmware on ESP32 microcontrollers, scalable Django/FastAPI backends containerised with Docker and orchestrated with Kubernetes concepts, and TensorFlow models in production.

My flagship project — Smart Vision Aid — is AI-powered smart glasses with 3× ESP32-CAM modules and YOLO object detection, giving the visually impaired 360° spatial awareness. It was selected for VIT Amaravati's Internal Expo 2025.

Most recently I built CyborgAI — a fully local, self-healing AGI OS submitted to the Kaggle × Google DeepMind Gemma 4 Good Hackathon, combining MedGemma 4B for on-device clinical reasoning, an adaptive education tutor, Mirofish knowledge graph, and Jarvis voice engine — all running offline on your own hardware with automatic health monitoring and failure recovery.

Alongside engineering, I run Editor Cyclops — a freelance video editing and content strategy practice serving 10+ creator-economy clients with short-form content for Instagram Reels and YouTube Shorts.

My superpower: ML theory → production systems. I understand the math, write the code, deploy the product — and cut the story together.