AI System
Real-Time Video Proctoring
Engineered exam monitoring with WebSocket, OpenCV & MediaPipe for reliable violation detection.
- WebSocket
- OpenCV
- MediaPipe
Hello, I'm
Soumika
Full Stack Developer</>
I build scalable web applications and delightful digital experiences with code, creativity and a lot of caffeine.
5+Projects
1+Years Experience
1000+Cups of Coffee
∞Passion
Tap to play
About
Nice to meet you
I'm Soumika, a Computer Science student and software engineer who loves turning fuzzy ideas into fast, delightful products. Currently pursuing my B.Tech at IEM Newtown, I specialize in full-stack development, backend systems, and AI/ML. When I'm not shipping code, you'll find me with a warm cup of coffee and a very opinionated cat supervising my commits.
Clean, typed, tested code
Design-minded engineering
Cat-approved commits
Skills
My toolbox
Frontend
- React
- Next.js
- Vite
- TypeScript
Backend
- Node.js
- ASP.NET Core
- FastAPI
- C#
- Python
Database & Ops
- PostgreSQL
- MySQL
- MongoDB
- Docker
- Linux
ML & AI
- TensorFlow
- PyTorch
- OpenCV
- NLP
- CNN/LSTM
Projects
Things I've built
NLP Tool
AI Resume Parser
Built an NLP parser extracting structured JSON from PDFs; reduced screening time by 8hrs/week.
- Python
- NLP
- JSON
Job Platform
Diaspora Job Marketplace
Launched a production job platform with JWT auth, OAuth, and payment integration.
- Next.js
- OAuth
- Payments
Research
Papers & Publications
Springer AISC SeriesIEEE Co-SponsoredJuly 24–26, 2025
Enhancing Brain Tumor Diagnosis with Ensemble Deep Learning Technique for Improved Accuracy
2nd International Conference on Artificial Intelligence and Sustainable Computing (AISC 2025)
Authors:
Apurba Nandi, Tista Mukherjee, Soumika Das, Nitu Saha, Sayantani Das, Sandip Mandal
Engineered an ensemble deep learning framework combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) networks across 3,200+ MRI scans. Evaluated high diagnostic accuracy and robustness for multi-class brain tumor classification while reducing single-model variance.
- 3,200+ MRI Scans
- CNN-LSTM Hybrid
- Ensemble Learning
- Springer AISC
- IEEE Kolkata Section
Experience
My journey
Oct 2025 — Present
Software Engineer Intern
PepHub
Engineered 3 full-stack platforms (Next.js, ASP.NET Core, Vite) with fast load times and reliable payments. Managed 3 production deployments.
Jun 2025 — Sep 2025
Software Engineer Intern
Cardiacrhythm (HealthTech)
Built WebSocket real-time ECG streaming infrastructure supporting 200+ concurrent streams with latency reduced to 180ms.
Jan 2025 — Jun 2025
Software Engineer Intern
RevMaxx (HealthTech)
Developed HIPAA-compliant medical transcription backend and built an AI microservice for medical entity extraction using NLP.
Dec 2024 — Dec 2025
Research Engineer Intern
IEDC, UEM Kolkata
Developed brain tumor classification via CNN-LSTM. Architected the "Satyameba" HPC cluster with Ansible automation.
Grab my resume
Want the full story? Download my CV for a detailed look at my experience, skills and the projects I'm most proud of.
Download CVContact
Let's build something
Have a project, a role, or just want to talk about cats and clean code? My inbox is always open.
dassoumika11@gmail.com