Final-year CS student who spends most of her time making models actually work in the real world — not just in notebooks.
I'm drawn to the messy, interesting gap between "it works on my machine" and "it works in production." That usually means a lot of PyTorch, Flask, Docker, and figuring out why the latency is 3x worse than expected.
More recently I've been getting into distributed backend systems — building event-driven pipelines with Kafka, caching with Redis, and putting Spring Boot to work for heavier backend lifting. It's a different muscle than ML work, but the obsession with performance and reliability is the same.
Lately I've been deep in computer vision (YOLO, OpenCV) and building backend systems that can serve ML predictions reliably. I care a lot about things like API design, response times, and not breaking things when you scale.
On the side, I keep sharpening my DSA and system design — partly because I enjoy it, mostly because I want to build things that don't fall apart.
Python Java JavaScript PyTorch YOLO OpenCV Flask Spring Boot Kafka Redis Apache MongoDB Docker Linux REST APIs
- Building event-driven and distributed systems alongside ML work
- Turning a few side projects into things I'd actually put in front of users
- GitHub: pragyabose1011
- LinkedIn: Pragya Bose