1st Year CSE @ MUJ | Building Computational Reality
I don't just build chatbots; I build autonomous systems that reason, act, and simulate reality. Currently bridging Scientific Discovery and Agentic AI.
| Project | Description | Tech Stack | Status |
|---|---|---|---|
| 🔬 VERITAS.AI | Physics-Based Deepfake Reality Engine | Gemini 3.0, Manim, SciPy | Building (Gemini 3) |
| 📈 SENTIENT110 | AI Financial Sentiment Analysis | Claude 3.5 Haiku, Story Protocol | FAIL.exe 2026 |
| 🤖 DugTrio | Verifiable Crypto Sentiment Agent | Solana, Story Protocol, RoBERTa | 🏆 Winner ($450 Prize) |
| 🛡️ Sentinel | Autonomous Financial Risk Agent with RAG | LangGraph, FastAPI, ChromaDB | ✅ Live |
| 🛡️ SHIELD | Multimodal AI Scam Detector | Azure AI, GPT-4o, Streamlit | Imagine Cup '26 |
- 🌟 Solana Foundation Audit Subsidy (2026): Selected for the Security Subsidy Program ($5,000 allocation) to audit smart contracts.
- 🏆 Winner, 100x Solana Hackathon — Top 10/232 & $450 Prize (Jan 2026)
- 🌟 Selected for WikiTech Cohort 2 — "Road to Wiki" Open Source Program (Top <1% applicants)
- 🥇 Synaptix AI Hackathon (IIT Madras Shaastra 2026) — Rank 14/2200+
- 🏅 Smart India Hackathon — Internal Round Winner
- ☁️ Google Cloud Skill Badges — 25+ certifications
Core AI & Agents
Full Stack & Systems
- Project VERITAS — Teaching AI to "see" physics in raw video and simulate it.
- Deep Think Models — Experimenting with reasoning chains in Gemini 3.
- Competitive Programming — Mastering C++ DSA.
"The best way to learn AI is to build production systems that actually work."