M.Tech in Computer Science & Engineering | IIT Kanpur (2022-2024)
B.Tech in Computer Science & Business Systems | Sister Nivedita University (2018-2022)
Building real-time voice AI systems with TTS, ASR, and streaming pipelines
Developed LLM applications, RAG systems, multi-language translators, and speech processing pipelines
🔊 DieTraAudio Processing Pipeline Denoising + Diarization + Transcription using DeepFilterNet, NVIDIA NeMo, and Mistral Voxtral. 🎯 10-12GB VRAM vs traditional 80GB+ |
7 Attention Mechanisms From vanilla to Flash & Multi-Latent Attention with complexity analysis. 📚 Deep dive into modern LLM architectures |
|
AI-powered MoM Generator Whisper + GPT-3.5 + PyAnnote diarization with React frontend. ⚡ Auto-generates professional meeting minutes |
RAG-based Conversational AI Context-aware chatbot with document retrieval and LangChain. 💬 Multi-turn conversations with memory |
🎨 More Projects
- 🎬 CrewAI YouTube Blogger - Multi-agent content creation
- 🚗 YOLOv8 Traffic Measurement - 97.5% mAP vehicle detection
- 📝 Text Summarization MLOps - Pegasus with MLOps
- 🎥 YouTube to Notes - Gemini-Pro summarization
- 🍷 Wine Quality ML - Complete pipeline with MLflow & AWS
- 👤 Face Recognition Attendance - Real-time tracking
- 🗣️ Voice Assistant - AI chatbot with audio capabilities
| Metric | Count |
|---|---|
| 💻 Commits | 0 |
| 🔀 Pull Requests | 0 |
| 👀 Code Reviews | 0 |
| 🐛 Issues Created | 0 |
Last updated: January 06, 2026 at 01:27 UTC
💼 Open to Voice AI, ML/DL roles and research collaborations
"Building real-time AI systems, one model at a time" 🎙️✨