AI Training for Enterprise Business Logic
Computer Code ProgrammingRLHF
This project involves fine-tuning a Large Language Model (LLM) to assist users in navigating complex enterprise resource planning (ERP) systems. My role focuses on RLHF (Reinforcement Learning from Human Feedback) and SFT (Supervised Fine-Tuning) to ensure the model provides accurate, compliant, and logically sound responses to functional business queries. I evaluate model-generated responses based on their technical accuracy regarding accounting principles, inventory valuation methods (such as FIFO and AVCO), and cross-functional workflow dependencies. By ranking multiple model outputs, I help the model learn to prioritize the most efficient and standard-compliant solutions for the end-user.
This project involves fine-tuning a Large Language Model (LLM) to assist users in navigating complex enterprise resource planning (ERP) systems. My role focuses on RLHF (Reinforcement Learning from Human Feedback) and SFT (Supervised Fine-Tuning) to ensure the model provides accurate, compliant, and logically sound responses to functional business queries. I evaluate model-generated responses based on their technical accuracy regarding accounting principles, inventory valuation methods (such as FIFO and AVCO), and cross-functional workflow dependencies. By ranking multiple model outputs, I help the model learn to prioritize the most efficient and standard-compliant solutions for the end-user.
2025 - Present