AI & Data Science Engineer Β· Full-Stack Developer Β· Cybersecurity Researcher
π Navi Mumbai, India Β· π« rohitswami537@gmail.com
Chameleon β a self-evolving, privacy-preserving honeypot system
PyTorch BiLSTM (~99.6% payload accuracy) + LoRA fine-tuned Qwen edge LLM via Apple MLX β ~99.97% combined detection rate
Two novel algorithms: TC-PSO (70% faster botnet tarpitting) Β· S-RRT (deceptive filesystem evolution under RAM constraints)
π Academic publication authorized β targeting IEEE ANTS / Springer ICISS 2026
- π₯ Best CTO β Bizence Startup & Problem-Solving Competition 2026
- π₯ Best Project in Fintech & Open Innovation β GDG AI Hackathon 2.0 (Mumbai Γ Pune)
- π Built an XGBoost + LSTM intraday trading model on Nifty 500 β 646.6% backtested return
- βοΈ Self-represented and won a consumer court case against Asus India Pvt Ltd
AI / ML
PyTorch TensorFlow Scikit-learn XGBoost LangChain LangGraph RAG LoRA Fine-Tuning Apple MLX NLP Computer Vision Synthetic Data Generation
Full-Stack
React.js Next.js Node.js FastAPI Express.js WebSockets TailwindCSS MongoDB PostgreSQL Firebase
DevOps & Cloud
Docker GitHub Actions GCP AWS (S3, Lambda) CI/CD
Blockchain
Solidity Ethereum Truffle Web3.js Ethers.js
| Project | Stack | What it does |
|---|---|---|
| Chameleon | PyTorch Β· MLX Β· Qwen Β· FastAPI | Self-evolving honeypot with novel PSO + RRT algorithms |
| FinManage | Next.js Β· Gemini API Β· Google ADK | Multi-agent AI FinTech platform β GDG Hackathon Winner |
| Chatty | MERN Β· Socket.io Β· JWT | Real-time chat β sub-100ms delivery |
| Trading Model | XGBoost Β· LSTM Β· GCP | Intraday signal model β 646.6% simulated return |
| Nyaya-Setu | FastAPI Β· LangChain Β· RAG | AI legal-tech for Indian consumer courts |
πΌ Open to remote part-time internships β AI/ML Β· Full-Stack Β· Cybersecurity Research

