Spaces:
Running
Real world use cases in the agents course
This is a discussion to talk about the use cases that we include in the agents course. So far, we're working on retrieval, code, and SQL.
🧑🎓 If you're a student, let us know what use cases you're interested in tackling.
👩🏫 If you're a pro, let us know what you want to contribute and your experience with that use case.
🙂
for multiple llm calls for chatbot, such as original question - > rephrase query -> gpt_response -> translate response as per original question language
for multiple llm calls for chatbot, such as original question - > rephrase query -> gpt_response -> translate response as per original question language
Nice detailed suggestion @jasondsouza09 . What kind of application would this be used in?
@burtenshaw I'm interested To solve a specific Data science use case, by building AI agents that need to interact with structured data, perform numerical calculations, or integrate external Python libraries like Auto EDA , specifically i'm interested in using PythonREPLTool of the LangChain framework for this use case .
I have been working on a blueprint for AI development, focusing on applied examples of RAG, information extraction, analysis and fine-tuning in the age of LLMs and agents.
Repository: https://github.com/huggingface/ai-blueprint
The idea of the project is to start with a realistic text dataset and solve a use-case. We will do this by creating various Gradio Spaces that will lead up to agentic approaches using Tool.from_spaces.
Current use cases, but open to suggestions:
- Agentic RAG
- Information extraction, labelling
@burtenshaw its a company's internal HR chatbot developed using aws kendra and openai