Summary
Kashyap Parmar is a data scientist and AI/ML engineer with a strong track record of shipping production-ready LLM and ML solutions across startups and enterprise projects. With an MTech in AI from IIT Hyderabad and eight years of engineering experience, he has fine-tuned and deployed custom Llama models, integrated multi-vendor LLMs (OpenAI, Bedrock, Hugging Face, Groq), and built RAG-powered chatbots and web extensions. He excels at end-to-end ML systems—data cleaning, labeling, model training, inference optimization, and cost/latency improvements—demonstrated by projects like a transliterated language detector that outperformed Google Translate and an insurance recommendation chatbot. Kashyap combines hands-on engineering (FastAPI, vLLM, LangChain, Pinecone/Chroma) with strategic product thinking, having advised on AI roadmaps and cross-functional impact. He often bridges research and product needs by applying advanced fine-tuning techniques (SFT, GRPO) and pragmatic deployment choices on AWS and edge runtimes. Based in Hyderabad, he brings a rare mix of academic rigor and startup velocity that accelerates ML projects from prototype to production.
7 years of coding experience
3 years of employment as a software developer
Master of Technology - MTech, Artificial Intelligence, 8.33/10.0 CGPA, Master of Technology - MTech, Artificial Intelligence, 8.33/10.0 CGPA at Indian Institute of Technology, Hyderabad
Gujarat Technological University
Kendriya Vidyalaya Sangathan