Building LLM Chatbots with Milvus to Leverage Your Internal Knowledge Base

September 28, 2023 3:00 PM
(ET)

Learn how to use Milvus — a fast, highly scalable open source vector database designed to provide relevant documents for a user’s query — with Shakudo to manage and get the most out of your company’s internal knowledge base.

Attendees will undergo a guided workshop to learn how to: 

  • Store their internal knowledge base data on Milvus
  • Use Milvus to perform fast vector searches on their data
  • Compute embeddings using an LLM embedding model
  • We will use MiniLM SBERT model in a Jupyter Notebook running in a Shakudo Session
  • Craft a prompt to send to a LLM endpoint for “speaking to their data”
  • The LLM we will use runs on Shakudo as a Service
  • Use the Shakudo platform, which provides an immediately available, fully managed version of Milvus and easily connects Milvus to other important stack components, including Nvidia Rapids and Dask

Attendees will undergo a guided workshop to learn how to: 

  • Store their internal knowledge base data on Milvus
  • Use Milvus to perform fast vector searches on their data
  • Compute embeddings using an LLM embedding model
  • We will use MiniLM SBERT model in a Jupyter Notebook running in a Shakudo Session
  • Craft a prompt to send to a LLM endpoint for “speaking to their data”
  • The LLM we will use runs on Shakudo as a Service
  • Use the Shakudo platform, which provides an immediately available, fully managed version of Milvus and easily connects Milvus to other important stack components, including Nvidia Rapids and Dask

Speakers

Jeremie Zumer
Jeremie is a Machine Learning Engineer at Shakudo, with experience developing deep learning techniques to solve bioinformatics problems. Previous work includes dialogue systems for automated agents at Microsoft and peptide identification by deep learning at IRIC.
Yujian Tang
Yujian is a Developer Advocate at Zilliz, the team behind Milvus, the world's most popular open-source vector database. During college, he interned at nCino and researched machine learning, publishing two papers. After joining Amazon, he contributed to generating $2.5MM with automated machine learning lifecycle systems, and since wears many more hats such as senior software engineer, full stack developer, technical evangelist, and blogger.