LLM Training

In training a Large Language Model, I was responsible for the model's performance through rigorous review and sanity checks. One of the key areas I specialized in was RLHF, short for Reinforcement Learning from Human Feedback. This involved refining the algorithms that enable large language models to learn from and adapt to feedback, optimizing existing code to ensure maximum efficiency. I also evaluated the quality of AI-generated code, by providing feedback to enhance response quality in English and Hindi.

I was also responsible for onboarding new team members and conducting comprehensive training sessions that covered the entire project workflow. These sessions weren't just about imparting knowledge but also included mentorship, both one-on-one and in group settings.

Throughout these responsibilities, my focus was always catering to the business and technical needs of the project.