Building real-time recommendation systems, forecasting tools, and production-grade data pipelines.
I'm a Computer Science postgraduate based in Auckland, New Zealand, working at the intersection of machine learning, data engineering, and real-world business problems.
My M.Tech thesis at NIT Karnataka (CGPA 8.36/10) focused on user interest drift detection in real-time streaming recommendation systems, combining CUDA-accelerated GPU inference, stratified reservoir sampling, and multi-view attention mechanisms.
Outside research, I build production tools: retail forecasting models, anomaly detection systems, and real-time data pipelines. I'm most energized by the moment a model stops being a notebook experiment and starts being something a business actually depends on.
Cloud platforms, Airflow, dbt, Kubernetes, and GitHub Actions reflect coursework, self-study, and project-level exposure rather than multi-year production ownership, happy to go deeper on any of these in conversation.
|
Real-time user interest drift detection using CUDA-accelerated GPU inference, stratified reservoir sampling, and multi-view attention.
|
Production tool flagging pricing irregularities in live retail data using statistical thresholding and a Flask REST API.
|
|
Forecasting model incorporating external signals (weather, election cycles) for retail demand prediction.
|
Mobile app using accelerometer-based collision detection with automatic GPS emergency alerting.
Published: IEEE ICCCMLA 2022
|
|
High-throughput event pipeline with live monitoring dashboard for real-time vote tracking.
|
Lightweight authentication system for industrial IoT device communication using socket-level cryptography.
|
| Title | Venue | Date |
|---|---|---|
| Efficient Parallel Algorithm for Detecting Longest Flow Paths in Flow Direction Grids | IEEE ISACC 2025 | Apr 2025 |
| Accident Rescuing System for Vehicles in Road Traffic: A Smart Phone Application | IEEE ICCCMLA 2022 | Dec 2022 |
