I build AI systems across Multilingual ASR, Voice AI, LLMs, and RAG pipelines, with a focus on research-driven, production-ready solutions.
- π Currently working on medical LLMs, multilingual speech systems, and scalable AI pipelines
- π Focused on Indian language AI, speech translation, speaker diarization, and LLM alignment
- π§ Experienced in SFT, DPO, RAG, Whisper fine-tuning, and real-time inference
- π B.Tech in Artificial Intelligence & Data Science
- βοΈ I also enjoy technical writing, mentoring, and building practical AI products
- Fine-tuned medical LLMs using SFT and DPO
- Built real-time multilingual ASR and translation pipelines using Whisper
- Worked on speaker diarization, speech-to-text, text-to-speech, and voice-preserving speech systems
- Developed RAG-based systems for long medical documents, summarization, QA, and document comparison
- Built scalable AI services using FastAPI, Flask, RabbitMQ, Redis, Docker, and TensorRT
AI/ML: PyTorch, TensorFlow, Transformers, Whisper, LangChain, LlamaIndex, RAG, SFT, DPO, NLP, Speech AI, LLMs
Infrastructure: FastAPI, Flask, Docker, Kubernetes, Redis, RabbitMQ, Kafka, TensorRT, MLflow
Tools: Git, GitHub, Postman, VS Code, Colab, Hugging Face