Description

Are you worn out by all the tedious mental effort of figuring out how to avoid dying of dysentery while on the Oregon trail? Sure, you could think through some simple heuristics or pay your nephew in skittles to do it for you, but that’s just so much work! I mean, it’s the AI future now and I was promised that I would no longer have to think. If you too find yourself in this situation, join us and learn how to build your own agentic RAG app to do all your Oregon trail thinking for you!

This interactive workshop will guide you step-by-step through the process of building an AI-enabled system, in Python, to answer the various make-or-break questions that come up along the trail. We will build an agent layer to respond to events that require branching logic such as deciding whether to continue on the trail or ask for more information. Additionally, we will also incorporate Retrieval Augmented Generation (RAG) in collaboration with our agentic layer for answering questions that require more specific context such as first hand data on how high the water can get in April before you can no longer cross the river.

After this session you will have learned how to:

• integrate LLMs as decision making agents to take different actions based on chat input

• set up and populate a vector database for use in Retrieval Augmented Generation

• combine agent, RAG, and prompting layers into a fully functioning chat based application

Notes

https://github.com/redis-developer/oregon-trail-agent-workshop

Transcript