I'm a dual-degree engineer (AI/ML at GGSIPU + Data Science at IIT Madras) who works across four areas depending on the problem: applied AI/ML, software engineering, data science, and front-end-heavy product builds. I'd rather ship a working system than stay in one lane — my projects below are tagged by what they actually demonstrate.
Currently exploring SDE, AI/ML, Data Science, and Frontend/UI-focused roles.
| Track | What I bring |
|---|---|
| AI / ML | RAG pipelines, prompt engineering, OpenAI APIs, imbalanced classification, LightGBM/XGBoost |
| SWE / SDE | Full-stack apps (React + Node/Python), REST APIs, Docker, CI/CD fundamentals, Git |
| Data Science | Feature engineering, statistical modeling, Pandas/NumPy, business analytics, dashboarding |
| Frontend / UI | React/Next.js, responsive & mobile-first builds, component-driven UI, role-based dashboards |
- Comment Category Prediction — Multiclass ML competition, ranked 388/2,744 (Top 15%), Macro F1 of 0.8627, on extreme class-imbalanced text data
- Finsights — Full-stack personal finance tracker with LLM-powered natural language queries over spending data
- DeVahan — Web3 vehicle ownership platform with role-based dashboards, smart contracts, and IPFS-backed metadata storage
- PramaanNFT — Smart contract-backed property validation system with NFT minting and PIN-based auth
AI / ML
Frontend
Backend / Data
Cloud / Web3
| Project | What it does | Tracks | Stack |
|---|---|---|---|
| Comment Category Prediction | Imbalanced multiclass text classifier, Top 15% / 2,744 teams, Macro F1 0.8627 | Data Science · AI/ML | LightGBM, XGBoost, TF-IDF |
| Finsights | LLM-powered finance tracker with NL query over spending data | AI/ML · SWE · Frontend | React, Node.js, Python, MongoDB |
| DeVahan | Web3 vehicle ownership with multilingual, role-based dashboards | SWE · Frontend | React, Solidity, IPFS, MetaMask |
| PramaanNFT | Property validation via NFT minting and smart contracts | SWE · AI/ML | React, Python, Solidity |
| Citizen Care Pharmacy | End-to-end business analytics on pharmacy sales data | Data Science | Python, Excel |
Hackathon winner (Zenith 4.0), badminton and cricket, and occasional video production. Open to SDE, AI/ML, Data Science, and Frontend roles — feel free to reach out.
Open to opportunities · reach out via email or LinkedIn above