• Full-Stack Developer
  • AI/ML Enthusiast
  • Research Author

ShourjyaChakraborty

I build full-stack products with machine learning inside them, and I care about every layer, from the model to the interface.

Selected work

04 projects
Baat2Badlav screenshot
01Civic intelligence

Baat2Badlav

A civic platform for India that turns citizen reports into signals for development planning. People describe a local problem by text or voice in English, Hindi or Bengali, Gemini structures it, and deterministic models group the reports into hotspots, development gaps and explainable priority scores.

React · TypeScript · Vite · Tailwind CSS · Gemini API · Express

02Voice AI support agent

ZentroMart VoiceOps

A voice customer-support agent for an Indian e-commerce marketplace, built for the IIT Guwahati × Gnani.ai VoiceOps Capstone 2026. It listens to a caller, answers from store policy with cited sources, verifies identity before sharing order details and escalates high-risk claims to a human team. It reached 100% decision accuracy on a 52-question development benchmark.

Python · Gemini · Gnani.ai speech · Hybrid RAG · FastEmbed · pytest

WhatNow? screenshot
03First-response guide

WhatNow?

A first-response guide for real-life emergencies such as a lost passport, a stolen card or a hacked account. You describe what happened in plain words, Gemini matches it to one of 52 human-audited scenarios, and you get a step-by-step plan with official helplines, deadlines and document templates for your country. It needs no account, and it keeps working without the AI.

Next.js · TypeScript · Tailwind CSS · Gemini API

BankLens screenshot
04Banking analytics

BankLens

A business intelligence dashboard for bank customer churn. Python and SQLite process a dataset of 10,000 customers, sort accounts into risk tiers and back each retention recommendation with the figure behind it. The front end is plain JavaScript with Chart.js, with live filters, spotlight search and CSV and PDF export.

Python · pandas · SQLite · JavaScript · Chart.js

About

Portrait of Shourjya Chakraborty

I’m a full-stack developer who works where web products meet machine learning.

Most of what I build starts with Next.js on the front and FastAPI behind it. When a project needs a model, I train it in TensorFlow or PyTorch and ship the whole thing on Vercel or Google Cloud.

Alongside product work I write research, compete in hackathons and practise algorithms, with 750+ problems solved so far. I like hackathons for the same reason I like shipping: a working thing in 48 hours beats a perfect plan.

Focus
Full-stack products with applied machine learning
Research
Published with Springer; First Best Paper Award at SCMIM 2025
Hackathons
Decisia 2025 winner, TechAI 2025 second place, XAVSPARK 2024 top 12 in India
Open source
Hacktoberfest contributor

Education

  1. 2023 – 2027

    Techno Main Salt Lake (MAKAUT)

    B.Tech in Computer Science & Engineering (AI & ML)

    Expected 2027

  2. 2021 – 2023

    Bhavans Gangabux Kanoria Vidyamandir, Kolkata

    Higher Secondary, CBSE

    91%

  3. 2012 – 2021

    Delhi Public School Megacity, Kolkata

    Secondary, CISCE

    96%

Toolkit

What I build with

Frontend

React/Next.js/TypeScript/Tailwind CSS/Framer Motion

Backend

FastAPI/Express/Supabase/Firebase/PostgreSQL

AI & ML

TensorFlow/PyTorch/Scikit-learn/Gemini API/Claude API

Cloud

Docker/Vercel/Google Cloud/Railway

Graphics

Three.js/React Three Fiber/WebGL/GLSL

Achievements

2024 – 2026
  1. 012026
    Semi-finalist

    Samsung Solve for Tomorrow 2026

    Top 40 IndiaTeam LeadTeam Valnex

    Led Team Valnex into the Top 40 teams nationally by building a mobile-app prototype for our proposed solution under the Health & Education domain. Translated the team’s problem statement into a tangible product concept, demonstrating hands-on development, initiative, and a commitment to technology-driven social impact.

  2. 022025
    Research award

    SCMIM 2025

    First Best Paper AwardCo-authored Research

    Received the First Best Paper Award for presenting our co-authored research paper, “Raga Reverie: A Raga-Origin Detection Algorithm and Its Therapeutic Implementations,” at the 1st International Conference on Sustainable Communication, Machine Intelligence and Metaverse, hosted by MCKV Institute of Engineering.

  3. 032025
    Published paper

    Springer Research Publication

    Published Research

    Our co-authored paper, “Raga Reverie: A Raga-Origin Detection Algorithm and Its Therapeutic Implementations,” was published in the Springer journal.

    View publication
  4. 042025
    1st place winner

    Decisia Hackathon 2025

    1st PlaceTeam Nexus Mavericks

    Secured 1st Place overall by building and demonstrating a solution in the Mental Health domain. Worked with the team to translate a sensitive real-world problem into a practical technology-driven solution, demonstrating rapid prototyping, problem-solving, and execution under hackathon constraints.

  5. 052025
    2nd place winner

    TechAI Hackathon 2025

    2nd PlaceTeam Nexus Mavericks

    Achieved 2nd Place by building and demonstrating an AI-focused solution in the Mental Health domain. Coordinated with team members to keep development aligned, tasks progressing, and deliverables completed smoothly and on time, demonstrating collaborative leadership, execution discipline, and the ability to deliver under pressure.

  6. 062024
    Top 12 finalist

    XAVSPARK 2024

    Top 12 IndiaTeam Nexus Mavericks

    Reached the final round of our first business hackathon, earning a place among the Top 12 teams in India.

Contact

Let’s talk.

I’m open to full-stack and applied machine learning work, research collaborations and a good hackathon team. Email is the fastest way to reach me.

sc.code05@gmail.com