Shanjal Gupta

Lead Ai Reasoning Engineer @Turing

Palo Alto, CA, US

Role & Responsibilities:Working on a project with one of the leading foundational LLM companies to enhance the performance of their Large Language Models (LLMs).Generating high-quality proprietary datasets for Supervised Fine-Tuning (SFT) and Reinforcement Learning from Human Feedback (RLHF) to improve LLM capabilities.Creating golden prompt-response pairs for SFT, leveraging open-source datasets and writing Salesforce code to address data analysis queries.Evaluating LLM-generated responses by comparing outputs from different model versions and providing critical feedback for model optimization.Assisting in benchmarking LLMs against competitor models by curating high-quality evaluation datasets.Collaborating with AI researchers and engineers to refine model training processes and enhance model performance.Key Contributions:Helped in the generation of proprietary data samples for fine-tuning and benchmarking LLMs.Provided feedback that directly influenced model improvements in areas such as reasoning, accuracy, and contextual understanding.Contributed to advancing AI by improving LLM training methodologies through high-quality data generation.

Cambridge Convent Senior Secondary School

Intermediate (10+2)

Graphic Era Deemed to be University

Bachelor of Technology - BTech