Making Cassandra Easier for AI Developers
summary
Cassandra becomes usable for AI by adding a document-layer HTTP API and vector features: a new Json API plus Storage Attached Indexes let Cassandra act like a document database and run vector search; client libraries for JavaScript (Mongoose/Astro DBTS), Python and Java plug in to LangChain and LlamaIndex. Examples include a photography app and a Taylor Swift chatbot that work in preview on DataStax Astra, often needing as little as five lines of JavaScript to fetch vectors and run RAG.
AI-written summary · about
chapters (9)
- 0:00Cassandra for AI developers
- 1:29Why Cassandra is hard
- 6:57Simplifying Cassandra with APIs
- 9:28New JSON API
- 17:16Vector search for AI
- 20:49AI apps with JSON API
- 25:02Language ecosystem support
- 30:00Cassandra and AI future
- 31:26Conclusion & Q&A
related talks
18mKeynote: Data = AI: Leading the Future of Gen AI with Apache Cassandra
33mUsing Apache Cassandra as your Document Database
50mHow to design and build AI applications with vector search using Azure OpenAI & Azure Cosmos DB
24mStaying Ahead of the Curve with Spring and Cassandra 4
12mBuilding Generative AI Apps with Azure Cosmos DB
14mBring AI to Your Data
20mBuilding AI Applications with Azure Cosmos DB JavaScript SDK
11mCassandra Vector Search Performance Tuning