Software Engineer
Prem Charan Pampana
Currently diving deeper into backend and distributed systems.
Chennai, Tamil Nadu, India
About
I'm interested in backend engineering, distributed systems, GPUs, Linux, rockets, and robots.
B.Tech in Computer Science Engineering, IIITDM JabalpurMadhya Pradesh, India · Dec 2021 – Aug 2025
Experience
Software Engineer
Cognizant Technology Solutions
2025 – Present
- Built a GenAI report generation pipeline that automatically produces site section reports, using Spring Boot, Angular, and Azure OpenAI REST APIs.
Backend Engineer
Fusion — Institute ERP
Dec 2023 – Aug 2024
- Architected 15+ RESTful APIs for primary healthcare features — appointments, inventory, announcements, medical relief workflows — cutting processing time by 40%.
- Built role-based access control across 4 user types: students, staff, dependents, and compounders.
- Designed scheduling and inventory modules covering 100+ beds, a blood bank, and 500+ medicines.
- Wired up report generation and dashboard logic with error handling and logging for day-to-day reliability.
Projects
Synchronous Chat Application
Real-time messaging app with direct and group channels, built on Socket.io.
- Bidirectional messaging with instant delivery and typing indicators, for both DMs and group channels.
- JWT auth over HTTP-only cookies with 256-bit encryption for session security.
- File sharing via Cloudinary — 10+ file types up to 25MB with real-time upload progress.
- 50+ reusable Tailwind components, responsive from desktop down to mobile.
ReactNode.jsExpressSocket.ioMongoDBTailwind CSS
Brainly — Your Second Brain
A second-brain app for organizing tweets, YouTube videos, and Google Docs in one place.
- React SPA with 10+ TypeScript components and an Express backend for saving and organizing embedded content.
- JWT-authenticated CRUD with a responsive masonry layout and real-time filtering.
- Shipped to production on Vercel/Render with MongoDB Atlas across 8 REST endpoints.
ReactTypeScriptNode.jsMongoDBTailwind CSS
Parallel MergeSort
A pthreads-based parallel merge sort, built to see how far thread-level parallelism could push a classic algorithm.
- 14x speedup over the sequential version on datasets from 10^6 to 10^8 elements.
- Mutex-based synchronization across threads, verified thread-safe over 50+ test cases.
- Tuned thread management and memory allocation to cut scheduling overhead.
C++pthreadsMultithreading
Stack
Languages
TypeScriptJavaScriptPythonSQLHTMLCSS
Frontend
ReactNext.jsshadcn/uiTailwind CSS
Backend & Database
PostgreSQLMongoDBMySQLNode.jsBunPrisma
Workflow & Tools
GitDocker
GitHub activity
Connect
Open to backend and full-stack roles, and to interesting problems in general.