Emil and Finn have built a mind-blowing Retrieval-Augmented LLM system called Kapa! In the podcast we discuss the E2E of building these systems from Data Ingestion, Cleaning, and Chunking to insights on how software companies (such as Weaviate haha) think about organizing their support tickets in these kinds of systems, as well as the latest trends in LLMs from Open-Source to Long Input Lengths and Fine-Tuning with LoRA! I hope you enjoy the podcast, more than happy to answer any questions or discuss any of your ideas about the content with you!
