As a model admin, I want to be able to collect the feedback on model responses and use it to direct model finetuning.

Consider and review suitable frameworks such as https://langfuse.com/ which provide standardized scoring and evalution of model responses. The content should be usable in a review process that would rely on instruction tuning (possibly using a low code approach such as InstructLab integrated in a User Interface)

As an expert user with limited technical knowledge, i want to be able to provide inputs in order to improve response of the model.

Evaluate the integration of RAFT (Retrieval Augmented Fine-Tuning) via InstructLab in the process of creating specialised agents for tuning.

@avijra @tuhinsharma121

Add deployment documentation and step--by-step for the deployment of the application on top of the GenAI LLM RAG pattern. Test and review the documentation via third-party to validate that new contributors can easily setup the app and get working.