Bridging analytics, LLMs and data products in a single database
In this webinar, we rethink how a modern, multimodal database system that uses arrays as its foundation can natively realize a data mesh that unifies tabular and complex data, generative AI, and data products. Sanjeev Mohan, Principal at SanjMo, and Stavros Papadopoulos, Founder and CEO at TileDB, engage in a thought-leading discussion on the future of databases.
The problem
Data Mesh and data products were the biggest trends that organizations were exploring to modernize their data analytics space… until generative AI crashed the party. Now organizations need to deal with more than relational analytics data, such as vector embeddings produced by large language models (LLMs) from a diverse spectrum of “unstructured” (i.e., complex, non-relational) data.
You know what would be the winning solution?
A database system that can support tables and SQL, complex data and vector embeddings, while enabling the design of data products built on this diverse data and associated workloads.
Contents of this video
0:00:00 – Welcome
0:02:20 – What is all the fuss about?
0:07:15 – The Data Mesh
0:13:42 – The Generative AI Hype
0:17:04 – What is an LLM?
0:20:50 – Customizing an LLM
0:27:28 – Similarity Search
0:30:22 – The rise of Vector DBs
0:41:42 – How does TileDB fit in the picture
0:52:12 – TileDB as a vector DB
0:56:07 – TileDB benefits for vector search
0:59:11 – TileDB as a data mesh
1:07:40 – Q&A
GitHub
https://github.com/TileDB-Inc/TileDB-...
About
TileDB is a multimodal database that molds to your application. Get amazing performance and minimize infrastructure.
Connect with us
Website: https://tiledb.com/
Twitter: / tiledb
LinkedIn: / tiledb-inc
Book a personalized product demo: https://tiledb.com/demo
Sign up at https://cloud.tiledb.com/auth/signup and contact [email protected] for free credits.