Final year B.Tech (AI & ML) student at Adani University (CGPA 8.24), graduating May 2027. First-author on an IEEE Xplore paper on agentic AI for text classification. Completed a research internship at IIT Patna, where I built an end-to-end CTG signal digitization pipeline, reconstructing fetal heart rate waveforms from clinical monitor images via pixel-calibration, then benchmarking classical ML vs. deep learning for fetal state classification. As Machine Learning Head at ASPDC, taught 150+ students end-to-end ML/DL pipelines and led a team of 5 ML Leads. Full detail in my resume →
- 🏆 Winner, Build with AI '26· Finalist, Tic-Tac-Toe '24 · Top 300, Amazon ML Challenge '24
- 📄 First Author, "Empowering Text Classification with Agentic AI: A Systematic Review" — IEEE Xplore
- 🎓 B.Tech CSE (AI & ML), Adani University - CGPA 8.24, Expected May 2027
| Category | Stack |
|---|---|
| Languages | |
| ML / DL | |
| GenAI / LLM |    |
| Backend / Web | |
| Databases / Vector Stores |   |
| Tools / Deploy |   |
| Project | Stack | What it does |
|---|---|---|
| TumorTrace | PyTorch, U-Net, FastAPI, React |
2D U-Net trained on 57K+ MRI slices (45K train / 11K val), >0.87 Dice on validation. Full-stack app with real-time tumor overlays, class-wise segmentation, auto-generated diagnostic reports. |
| LekhAI | Flask, Gemini, DuckDuckGo |
Multi-agent pipeline (Researcher → Strategist → Writer → Editor) with live web retrieval; automates long-form blog drafting end-to-end. |
| Sarathi | Gemini, React, TypeScript, Flask |
RAG system with vector search for citation-backed, context-aware retrieval over curated scripture data. Deployed end-to-end. |
| Swabhav | TensorFlow, Scikit-learn |
DNN classifier trained on 43K+ samples, 89.06% test accuracy; model selection driven by F1-score and overfitting-gap analysis, not just accuracy. |
-
Research Intern, Indian Institute of Technology - Patna ·
22nd May 2026 - 31st July 2026Designed a CTG digitization pipeline extracting fetal heart rate waveforms from clinical monitor images via pixel-calibration, improved robustness to glare, reflections, and incomplete grid structures. Engineered physiological features and benchmarked ML and DL for fetal state classification.
-
Machine Learning Head, ASPDC ·
Aug 2025 – Aug 2026Led a team of 5 ML Leads to design and deliver ML initiatives, taught 150+ students across "ML in One Month" and "DL in One Month" covering full pipelines from data collection through ANN/CNN/RNN/LSTM to deployment.
-
Machine Learning Lead, ASPDC ·
Aug 2024 – Aug 2025Ran ML-101 workshops introducing linear regression and supervised learning fundamentals.