Nikhil Raj

1. Built a real-time video streaming and two-way audio platform handling 2.5+ TB/month, enabling reliable, low-latency communication at scale.
2. Designed scalable pipelines to ingest and process binary IoT streams, improving real-time data delivery and overall system throughput.
3. Developed ML validation pipelines to reduce false-positive alerts (e.g., smoking, seatbelt), processing 30K+ images/hour and improving alert accuracy.
4. Engineered a high-throughput geofencing system processing 15M+ events/day, using PostGIS for real-time spatial queries and ScyllaDB for low-latency alert delivery.
5. Built a distributed image-to-video processing pipeline handling 200K+ images/day, optimising performance through parallel and asynchronous execution.

• Contributed to building the backend for a no-code data analytic platform, leveraging Python, Fast API for developing REST APIs, MongoDB as a database, and Redis for caching.

• Developed a robust OTP authentication system for secure user authorization and implemented in-app, Slack, and email notifications using Python and FastAPI.

• Transformed data from MongoDB and other databases into structured formats, creating diverse visualizations like heatmaps, box plots, and radar charts to effectively represent complex data.

• Worked on web-assembly-based concepts. • Built different Python utility tools for document and image processing. • Integrated these tools into the website using required libraries implemented in Next.js and Typescript.