PROJECT |
WHAT IT IS |
BUILT WITH |
|---|---|---|
| MindFirst | Mental-health triage — AI chatbot + psychometric screening in one platform | Next.js Flask ML |
| IntelliStatement | Bank statements in, financial insight out — LLM extraction from PDFs & images | FastAPI Streamlit LLM |
| Rail Mitra | Railway platform & track assignment, solved as constraint programming | Python OR-Tools |
| PixelForge | Leukemia image classification with CNNs, GAN-augmented on C-NMC | PyTorch GANs |
| Sim Sim | Voice-driven desktop assistant with a native GUI | Python PySide6 |
| Committee Connect | Task & attendance tracking for college committees | Android Firebase |
The rest — ArtScape, To-Do, and the site itself — lives at mosambiswas.com.
Comparative Analysis of Psychometric Data for Mental Health Assessment — ICAIIHI-2025. PHQ-9 / GAD-7 psychometrics through five supervised models at ~99% accuracy, kept interpretable with SHAP — a smarter first step for mental-health triage.
Python Java JavaScript C++ — PyTorch Next.js Flask FastAPI Node — Firebase MongoDB Android Git
Currently deep in: LLMs · computer vision · system design
Street, portraits, monochrome, concerts, wildlife. Technical Head, Photocircle RAIT ’22–’25. The archive is SheiChobi — Bengali for “that picture.”
© 2026 MOSAM BISWAS — NAVI MUMBAI, IN · NO TEMPLATES, NO WIDGETS — BUILT BY HAND
