Spinning up on Postgres & AI with Arda Aytekin | Path To Citus Con Ep13

Everywhere you look, people are talking about AI. From Copilot to ChatGPT to Postgres’s powerful AI capabilities (think: pgvector), AI is everywhere. In this episode of Path To Citus Con*, for developers who love Postgres, Arda Aytekin joined Claire and Pino to chat about spinning up on Postgres and AI. Arda shared his origin story in mechanical engineering and data science before walking us through vector databases, pgvector, and azure_ai. Arda is one of the creators of the azure_ai extension, so the conversation delves into the azure_ai integration between Azure Database for PostgreSQL and Azure AI Services. Also discussed (of course) was-Responsible AI.
* Path To Citus Con has been renamed to Talking Postgres. Past podcast episodes can be found here: talkingpostgres.com
Guest Arda Aytekin:
Arda is a senior software engineer on the PostgreSQL team at Microsoft where he has taken an active role in the AI/ML-related efforts on and around Azure Database for PostgreSQL. Arda earned a Ph.D. in asynchronous first-order algorithms for large-scale optimization problems and is a self-described open-source enthusiast.
Co-hosted by Claire Giordano and Pino de Candia:
Claire Giordano is head of the Postgres and Citus open source community initiatives at Microsoft. Prior to the Microsoft acquisition of Citus Data, Claire served in leadership roles in engineering, product management, and product marketing at Sun Microsystems, Amazon/A9, and Citus Data.
Pino de Candia is a software dev manager at Microsoft since 2020 and is currently working on the Citus open source project. Pino previously worked on the managed PostgreSQL database service in Azure Cosmos DB for PostgreSQL, which includes Citus on Azure support for distributed PostgreSQL. Pino has lived in New Orleans since 2017.
Chapters:
⏩ 1:45 Arda’s origin story
⏩ 11:11 What is pgvector?
⏩ 19:16 azure_ai extension to Postgres
⏩ 30:22 Responsible AI
⏩ 44:07 Text embeddings
⏩ 50:31 Tips on spinning up on Postgres & AI
📜 Full transcript of this podcast episode available at:
pathtocituscon.transistor.fm/...
✅ Listen to more episodes of Path To Citus Con:
aka.ms/PathToCitusCon
📅 Subscribe to the monthly Path To Citus Con calendar:
aka.ms/PathToCitusCon-cal
Links mentioned in this episode:
🔹pgvector: github.com/pgvector/pgvector
🔹Andrew Kane: github.com/ankane
🔹Simon Willison’s Blog: simonwillison.net/
🔹Demo of Azure AI & pgvector with Azure Database for PostgreSQL by Claire Giordano: • Demo of Azure AI & pgv...
🔹Introducing the azure_ai extension to Azure Database for PostgreSQL by Denzil Ribeiro: techcommunity.microsoft.com/t...
🔹Azure AI Extension, Microsoft Learn: learn.microsoft.com/azure/pos...
🔹Vectors are the new JSON in PostgreSQL by Jonathan Katz: jkatz05.com/post/postgres/vec...
🔹Responsible AI at Microsoft: aka.ms/rai
🔹Andreessen's Corollary: Ethical Dilemmas in Software Engineering by Bryan Cantrill: • Andreessen's Corollary...
🔹Phi-2: The surprising power of small language models by Mojan Javaheripi & Sébastien Bubeck: www.microsoft.com/research/bl...
🔹pg_vectorize: github.com/tembo-io/pg_vectorize
🔹OpenAI API: platform.openai.com/docs/intr...
🔹Microsoft Azure AI Fundamentals: Generative AI Training: learn.microsoft.com/training/...
🔹ChatGPT Prompt Engineering for Developers: www.deeplearning.ai/short-cou...
🔹Andrej Karpathy's keynote @ Microsoft Build 2023: • State of GPT | BRK216HFS
🔹Stanford University CS231n: Deep Learning for Computer Vision: cs231n.stanford.edu/
🔹LangChain: www.langchain.com/
🔹Towards Data Science: towardsdatascience.com/
🔹Generative AI for Beginners: github.com/microsoft/generati...
🔹Zero-shot learning: en.wikipedia.org/wiki/Zero-sh...
🔹Path to Citus Con Ep01: Working in public on open source with Simon Willison & Marco Slot: pathtocituscon.transistor.fm/...
🔹The Art of PostgreSQL by Dimitri Fontaine, get 15% OFF with CLAIRE15: theartofpostgresql.com/
🔹Path to Citus Con Ep08: Solving every data problem in SQL w/Dimitri Fontaine & Vik Fearing: pathtocituscon.transistor.fm/...)
🔹Arda Aytekin at PGDay Chicago 2024 on April 26: postgresql.us/events/pgdaychi...
🔹CFP for POSETTE open until April 7th 2024: aka.ms/posette-cfp-2024

[@rosendo3219]

if we could lower selectivity on the slides in this presentation, would be awesome! otherwise, great content as always.