Building data systems that see, stream, and scale.
I engineer production ML pipelines, real-time data platforms, and developer tools. Most of my work sits at the intersection of computer vision, streaming architectures, and applied AI — with a soft spot for sports analytics.
fifa-soccer-ds — Soccer Video Analysis Pipeline
End-to-end CV system for real-time soccer analysis. Fine-tuned YOLOv8 for player/ball/referee detection, ByteTrack for multi-object tracking, and GraphSAGE neural networks for tactical pattern recognition. Served via FastAPI with MLflow experiment tracking.
YOLOv8 ByteTrack GraphSAGE FastAPI MLflow DVC Docker
soccer-vision-research — Vision Model Benchmarks for Sports
Benchmarking RF-DETR, SAM2, SigLIP, and YOLOv8 architectures for soccer player detection and segmentation. Includes reproducible experiments and a live demo.
PyTorch RF-DETR SAM2 SigLIP W&B
contextbox — AI-Powered Context Capture
Developer tool for capturing screenshots, extracting web content, and querying your work context with semantic search. Supports multiple LLM backends with zero API cost via GitHub Models.
Python LLMs OCR Semantic Search FastAPI
voxt — Voice-to-Text Clipboard Tool
Linux CLI tool written in Go. Global hotkey triggers recording, Groq Whisper transcribes, result lands in your clipboard. System tray integration, SQLite history, Wayland + X11 support.
Go Whisper PortAudio DBus SQLite
stock-data-platform — Real-Time Stock Data Pipeline
Streaming data platform with Kafka producers/consumers, PostgreSQL persistence, and Airflow DAGs for orchestration. Fully containerized with Docker Compose.
Kafka Airflow PostgreSQL Docker Python
Languages
ML & Vision
Data Engineering
Infrastructure
Currently open to: Data Engineering, ML Infrastructure, and Backend roles — remote or India-based.
