analiza aiqbench
This presentation details how PostgreSQL serves as a critical infrastructure layer for scaling production-grade agentic workflows and LLM applications through advanced vector indexing and multi-modal capabilities. By leveraging features like JSONB parsing and high-recall HNSW, teams can achieve deterministic semantic context retrieval essential for enterprise reliability. The discussion highlights specific strategies such as vector quantization to optimize query performance while managing complex agentic memory states.
Presentation: Postgres for Production Agents: Your Relational Foundation for Enterprise AI
Gwen Shapira shares how teams are scaling AI features using PostgreSQL for mission-critical apps. She explains how to leverage Postgres's multi-modal capabilities - including JSONB parsing and high-recall HNSW vector indexing - to deliver deterministic and semantic context to LLMs. She also discusses vector quantization to speed up queries by 4x and strategies for managing agentic memory. By Gwen Shapira
Gwen Shapira: I'm here to talk about how Postgres is being used to drive AI features, mostly in large scale and where high reliability is critical. I am a big fan of relational databases in general, and Postgres in particular. I love the code. I love the community. I love it so much I founded a Postgres company. Our customers, especially in the last year, a lot of them are doing AI. We're internally developing some AI features. We have a lot of lessons to share. This is what I'm about to do. Usually when I give this type of talks, the audience has two parts to it. Those who know Postgres or some relational databases really well and are interested in the AI stuff, and those who actually know AI very well on other platforms and are more interested in the Postgres stuff. I have something for everyone. On the other hand, half of the audience may be bored at any specific point in time. I'm going to start in a bit of a roundabout way. We will get to the point soon enough. In 2023, ChatGPT was announced. The entire world exploded. Everything changed forever. At this point…
Extras afișat conform termenilor RSS ai publicației · 200 cuvinte · textul integral și figurile sunt pe site-ul publicației
Subiecte conexe
[AINews] Reality Checks on AI News (Yegge shuts down Gas Town, Databricks’ +60% Astra cost)
Latent Space · 17 sept. 2026
Netskope (NTSK) Q2 2027 Earnings Call Transcript
fool.com · 10 sept. 2026
[AINews] Meta Muse Wonder & Wonder Full: MSL’s Claude 5.5-tier Frontier Model
Latent Space · 13 oct. 2026
State of the Field: AI for Epistemics and Coordination
LessWrong · 2 oct. 2026