computer vision, edge AI engineer + ML developer with a passion for high-frequency charts & neo-retro bikes
I optimize real-time deep learning pipelines and deploy computer vision systems on low-power edge hardware. The objective is to design hyper-lightweight, high-throughput model architectures running at maximum physical frame rates. 🦀⚙️
Detailed writeups on optimizing models, Quantization-Aware Training (QAT), and TensorRT execution are hosted directly on my repositories.
| Role | Company/Institution | Location | Dates |
|---|---|---|---|
| Data Science Intern | IIT Madras | Chennai, India | Dec 2024 - May 2025 |
| AI Development Intern | SkyGad | Remote | Aug 2024 - Dec 2024 |
- ⚡ Real-Time Traffic & Surveillance Pipeline (IIT Madras): Deployed directional-aware YOLOv5/v8 object detection on NVIDIA Jetson systems. Compressed layers via TensorRT to maintain high real-time throughput. Integrated Annoy-based access tracking and automated reporting mechanisms.
- 👁️ Face Recognition & Temporal Tracking System: Real-time facial validation linking YOLOv8, ByteTrack, and ArcFace embeddings queried against a dynamic FAISS index. Stabilized with temporal label smoothing across video sequences.
- 🤖 RAVA (Retrieval-Augmented Virtual Assistant): Memory-enabled conversational agent utilizing LangGraph, Google Gemini API, and Annoy index indexing for per-user localized context.
- 📐 NumPy Feedforward Neural Network (CS6910): Built a deep feedforward network from absolute scratch (zero PyTorch/TF autodiff). Custom-implemented 6 optimizers (SGD, Adam, Nadam, RMSProp) and used WandB sweeps to benchmark Fashion-MNIST accuracy.
- Google Developer Student Club (GDSC) IIT Madras — Core Team Member & Technical Speaker ("Git and GitHub", "Dumping of Windows").
- Linux Community Lead — IIT Madras.
- Al Horizons Conference — Co-organized academic AI horizons forum alongside Prof. Sudarshan Iyengar.
- Smart India Hackathon — Runner-up (2023).