Debalina Biswas

About

Fourth-year Artificial Intelligence and Machine Learning student with strong problem-solving skills and hands-on experience in full-stack development and AI-based applications. Passionate about building impactful, real-world solutions using computer vision, machine learning, and scalable web technologies.

Education

B.E.

BNM Institute of Technology, Bengaluru

January 2023 - Present

Field: Artificial Intelligence & Machine Learning

12th

St. Stephen’s School, Kolkata

January 2020 - January 2022

Field: ISC, PCM

10th

St. Stephen’s School, Kolkata

January 2019 - January 2020

Field: ICSE

Skills

Intermediate

C CSS Data Structures and Algorithms Deep Learning Express.js Git GitHub HTML Java JavaScript MongoDB MySQL Node.js NumPy Object-Oriented Programming Pandas Python React.js Scikit-learn TensorFlow

Projects

KrishiSetu AI – Agentic AI Platform for Crop Intelligence

April 2026 - Present

Developed an agentic AI system to provide crop recommendation, market price insights, and farming advisory services. Built modular agricultural intelligence tools for fertilizer guidance, pest management, irrigation advice, and weather-based recommendations. Designed a decision-support system aligned with SDG 2 (Zero Hunger) to improve productivity and enable data-driven farming decisions.

KENDRA – AI-Powered Productivity Assistant

February 2026 - April 2026

Developed a Chrome extension to track browsing behavior and monitor website usage in real time. Built a Flask backend with REST APIs and SQLite database for storing and processing user activity data. Implemented website time-limit alerts and intervention-based blocking for distracting platforms. Integrated the Mistral AI LLM via LangChain and API calls to generate personalized productivity insights and behavioral intervention prompts.

Sign Language Recognition Web Application

February 2025 - March 2025

Developed a full-stack web application for real-time sign language recognition using computer vision and deep learning. Implemented hand landmark detection with MediaPipe and gesture classification using TensorFlow. Integrated OpenCV for real-time webcam capture and gesture preprocessing. Designed responsive frontend interface providing instant feedback for interactive learning.