7. Designing/Building Our Python FastAPI Layer
What piece comes first, our SpringBoot layer or conversational RAG? Really quickly, before we get into today’s blog post - I do want to say thank you. I hope the blog has been as exciting for you ...
Eddie Chan
6. Ingesting our Slack Data Source
Another day, another dolla’! We are another step closer to integrating slack messages as an additional data source. I have: Implemented chunking and embedding for our slack messages reading the...
5. Exploring Additional Data Sources
Exploring Adding Additional Data Sources to the Context Pool In today’s episode: slack messages! With a good bit of the foundational work out of the way, I think it’s a good idea to add on additi...
4. Designing the Hybrid Search, and problems continued...
Designing the Hybrid Search Approach Diving into the flow for the hybrid search, here’s what I’m thinking in terms of the workflow: Embed The query (no changes here) Run the vector search (t...
3. Learning About Hybrid Search and Modifying Data Stores
Learning about Hybrid Search Todays focus: Adding hybrid search want to attempt combining vector search (currently implemented) and keyword search for example if we search specific s...
2. Building A Roadmap
Building a Roadmap Next Steps: Roadmap proposed by Claude: Step Title Description Step 1 Upgrading Our Datastore Swap index.pkl for p...
1. Building a RAG system From Scratch
Experimenting with PDF file consumption Experimented with reading PDF files using pymupdf, library had a hard time reading tables content wasn’t lost, but columns were jumbled around. Coul...