Hi, I'm Saumyadeep 👋

Data Science specialist with experience in Machine Learning, Deep Learning, and Full-Stack Development. Passionate about applying AI to solve real-world problems.

SM

About

I hold Bachelor of Science in Data Science from CHRIST UNIVERSITY, graduated in June 2024. I have experience in various areas of data science and software development, including machine learning, deep learning, and full-stack web development. I've completed internships at the Indian Space Research Organization and Nehru Arts and Science College(in collaboration with ICAR), where I worked on projects involving National Health Resource Repository data analysis and disease detection using deep learning.

Skills

Python
Java
JavaScript
TypeScript
React
Node.js
SQL
Machine Learning
Deep Learning
AWS
Docker
Kubernetes
My Projects

Check out my latest work

Here are some of the projects I've worked on during my academic and professional journey.

Interactive Web Interface for LLM Models

Developed a ChatGPT-like web interface using React and Node.js, integrating 8 LLM models via Ollama. Implemented real-time response streaming, reducing average response time by 30%. Optimized load time by 40% and reduced server resource usage by 25% via efficient Docker deployment.

React
Node.js
Docker
Kubernetes

Traffic Light Control Optimization

Developed a Reinforcement Learning model to optimize traffic light timing, reducing average wait times by 27% in simulated urban environments. Improved traffic flow efficiency by 35% during peak hours. Implemented a Deep Q-Network (DQN) algorithm, training it on 100+ simulation episodes.

Python
PyTorch
SUMO
Reinforcement Learning

Automobiles E-Commerce Website

Designed and developed a full-stack platform for vehicle sales, enabling users to browse, compare, and purchase over 500+ car models. Implemented secure payment integration and achieved a 30% improvement in page load times.

Node.js
React
PostgreSQL
Redis

Custom QA Chatbot Development

Developed an advanced NLP-based chatbot that answers queries in 7 domains (Science, Technology, History, Geography, News, Sports, etc.). Increased user engagement by 30% through personalized recommendations and real-time global event updates. Achieved 95% accuracy in query understanding using NLTK for text processing and a sequential model with 3 dense layers. Optimized performance with Stochastic Gradient Descent (SGD) and softmax activation, refining response accuracy by 20%.

Python
NLTK
Machine Learning
Deep Learning

Basic Web Browser Development

Developed a basic web browser using Python and PyQt5, featuring a user-friendly interface with support for 5+ tabs. Implemented key navigation functions, enhancing user efficiency by 40%. Designed an intelligent URL bar and SSL status indication, providing improved accessibility and security awareness.

Python
PyQt5
HTML
CSS
Contact

Get in Touch

Feel free to reach out to me at saumyadeepmitra.12@gmail.com or connect with me on LinkedIn. I'm always open to discussing new opportunities and collaborations.