Summary
Pranjal Doshi is a Software Engineer with 9 years of industry experience, currently building advanced dialog management systems at NVIDIA that bridge speech, NLP models, and LLMs for multi-user, multi-turn interactions. He has hands-on expertise fine-tuning domain classifiers and intent-slot models, creating datasets, and deploying edge-capable solutions on Jetson for real-time natural language processing. Pranjal designs safe, microservice-based integrations that let third-party Python apps interact with complex conversational frameworks while enforcing validation rules to prevent data leaks and code injection. His work spans end-to-end delivery—unit testing, evaluation pipelines, and supporting multiple virtual assistants used internally—demonstrating an emphasis on reliability in production. Trained at IIT Kharagpur (MTech) and with a background in full-stack conversational tooling, he blends research-grade machine learning with practical engineering. A less obvious strength is his focus on secure, auditable interactions between LLMs and external code, enabling powerful extensibility without sacrificing safety.
8 years of coding experience
Bachelor of Engineering - BE, Computer Science, Bachelor of Engineering - BE, Computer Science at Shri Vaishnav Institute of Technology and Science, Indore
Master of Technology - MTech, Computer Science, Master of Technology - MTech, Computer Science at Indian Institute of Technology, Kharagpur