Hi HN, I started working on Context Data because I wanted to build a super simple infrastructure to build end-to-end data pipelines for my search and RAG projects especially ones that required vector databases.
Essentially, I set out to build the “FiveTran for Generative AI” where you can connect to multiple data sources, embed the data using all major embedding models and write the results to your vector database targets without having to write and rewrite tons of code.
You can also use SQL CTEs for transformations and aggregations as well as scheduling your end-to-end pipelines.
Essentially, I set out to build the “FiveTran for Generative AI” where you can connect to multiple data sources, embed the data using all major embedding models and write the results to your vector database targets without having to write and rewrite tons of code.
You can also use SQL CTEs for transformations and aggregations as well as scheduling your end-to-end pipelines.