ML Engineer from NSUT, New Delhi with a strong academic background, skilled in LLMs, optimizing ML models, and solving real-world software engineering problems. Currently exploring multi-agent systems and adversarial self-play for AI capability optimization.
- Kaggle Master | Amazon ML Challenge National Winner
- Currently working on Agent-groupStudy -- adversarial self-play for multi-objective AI agents
- Contributing to LiveKit Agents -- Realtime voice AI agents framework
- Building and working on Voice AI Agents
- Built satellite ground station GUI for CanSat Competition (GUI-Cansat-22)
| Project | Description | Tech |
|---|---|---|
| Agent-groupStudy | Adversarial self-play for multi-objective capability optimization | Python, LLMs |
| LiveKit Agents | Contributed Sarvam AI plugin addon for realtime voice agents | Python |
| NannyML | Contributing to ML monitoring & model performance tracking | Python, ML |
| GUI-Cansat-22 | Ground station GUI for a Can-Sized Satellite | Python, PyQt |
| credit-ease | AI-based credit scoring without credit history | Python, ML |
| Dataset | Downloads | Views | Upvotes |
|---|---|---|---|
| Global Data on Sustainable Energy (2000-2020) | 28,897 | 137,827 | 279 |
| Global Food Price Inflation | 15,591 | 88,686 | 206 |
| Notebook | Topic |
|---|---|
| Classify and Create Your Own Image Dataset | Computer Vision, Data Curation |
| Global YouTube Statistics: Interactive Auto EDA | Auto EDA, Data Visualization |
| Article | Claps |
|---|---|
| Craft a Powerful Reward Function in Student League (AWS DeepRacer) | 336 |
| Why You Need to Babysit ML Models After Deployment | 138 |
