Talking Postgres with Claire Giordano

What’s your gold standard for deciding whether AI-written code is ready for production? In Episode 42 of Talking Postgres, Simon Willison—creator of Datasette, co-creator of Django, and prolific open-source developer—joins Claire to share how AI is changing the way he builds software today. We dig into why his test for shipping AI-generated code is “Could I explain this to somebody else?”, how engineering management skills can be surprisingly useful when managing AI agents, and how the bottleneck is no longer writing code but understanding it. Also: personal credibility, deep research, the Winchester Mystery House, slop proxies, and Simon’s observation that “Features are cheap. That doesn’t mean you should build them all.”

Previously on Talking Postgres:
Talking Postgres podcast Ep 30: AI for data engineers with Simon Willison 
Links mentioned in this episode:
Blog: Simon Willison’s Blog
Project page: Datasette, for finding stories in data
GitHub repo: sqlite-utils
Blog post: Release of alchemy-utils 0.1a0 (featuring Postgres & DuckDB)
Blog post: There are no lossless transformations of natural-language text, by Sophie Alpert
Podcast episode: The Open Weight Revolution with Simon Willison, on Oxide and Friends
Podcast episode: An AI state of the union, on Lenny’s Podcast 
Wikipedia: Winchester Mystery House
Idea in Mythical Man-Month: Conceptual integrity
Blog post: SQLite compressed text-history prototypes, by Simon Willison
Tools: Simon’s miscellaneous tools repo
In this episode, we covered:

00:00 Intro & music
04:00 A week of work before breakfast
09:12 Red-green TDD makes agents exercise every line
14:42 A million lines mean nothing if you don’t understand it
18:28 Finding low-hanging fruit among open PRs & issues
22:07 Getting work done while walking the dog
23:01 Gold standard: “Could I explain this to somebody else?”
28:27 Slop proxies add no value at all
30:49 Aggressive nitpicking reviews
35:01 Everything in software engineering is about trade-offs
41:16 Racoon heist game is only fun for 1 minute 15 seconds
43:46 New Year’s resolution to be more ambitious
48:11 You have to learn to throw things away
51:57 Features are cheap. That doesn’t mean you should build them all
1:00:45 Engineering management skills are so useful
1:04:44 QA specialists should be having a great time
1:07:53 Writing is thinking, don’t outsource it
1:08:23 Skill atrophy is a choice you make
1:10:07 I believe in people who are motivated
1:11:55 Systems are not set up to deal with this volume
1:18:48 Research agents stopped being absolute garbage
1:21:49 The whole point of the “human in the loop”
1:26:10 Dream: I want there to be more small businesses

What happens when a database engineer spends 13 years on SQL Server and then starts working on Postgres? In Episode 41 of Talking Postgres, Distinguished Engineer Panagiotis “Panos” Antonopoulos of Microsoft joins Claire to share what surprised him most about making that transition. We explore why Postgres felt familiar from the start—and what shocked Panos about the clean and lean Postgres codebase. We also dig into how public mailing lists and AI can help explain why design decisions were made, why more people are asking “why not Postgres?”, and the shared-storage architecture behind Azure HorizonDB.

Previously on Talking Postgres:
Talking Postgres Ep 39: From MemSQL to HorizonDB, an engineer's journey with Adam Prout
Talking Postgres Ep 38: How I went from Oracle to Postgres (with a big NoSQL detour) with Gwen Shapira
Talking Postgres Ep 40: How I got started running a Postgres user group with Jeremy Schneider
Links mentioned in this episode:
SIGMOD: ACM SIGMOD-PODS Conference 2027  
CMUDB talk: HorizonDB: Co-Designing Postgres and Azure for Cloud-Native OLTP, by Adam Prout
PostgreSQL mailing list: pgsql-hackers
PostgreSQL docs: pg_stat_slru (simple least-recently-used)
Docs: Azure HorizonDB
Poster: Postgres Happiness Hints

Intensely local user groups have been part of Jeremy Schneider's story from the start—from Linux meetups at a Michigan coffee shop to a closet server running an Oracle database nobody knew anything about. In Episode 40 of Talking Postgres, Postgres engineer and Seattle Postgres User Group co-organizer Jeremy Schneider joins Claire to share how community led him to Postgres after 15 years with Oracle—and why "it's like I was born to be here." Plus: the newly-updated Postgres Happiness Hints poster, advice for starting your own user group, and his POSETTE 2026 talk on CloudNativePG.
Previously on Talking Postgres:
Talking Postgres Ep 38: How I went from Oracle to Postgres (with a big NoSQL detour) with Gwen Shapira
Links mentioned in this episode:
Seattle Postgres User Group: Meetup page
Seattle Postgres User Group: YouTube channel
User Group Map from PGConfEU 2025 talk: 48 Postgres User Groups during PG18 timeframe
PostgreSQL.org: Listing of local Postgres User Groups
Postgres Meetup For All (a virtual meetup): Meetup page
Jeremy Schneider’s Blog: Ardent Performance Computing
Poster: Postgres Happiness Hints
PGConf.dev 2026: Posters from Poster Session
PGConf.dev 2026 Poster Session: Talking Postgres poster
POSETTE: An Event for Postgres 2026: Jeremy’s POSETTE 2026 talk with Leonardo Cecchi
POSETTE 2026: Livestream 3 schedule & talks
Oracle docs: Oracle Database Concepts PDF
Book: Oracle Insights: Tales of the Oak Table

What does it take to make Postgres and Azure fit together cleanly, like puzzle pieces? In Episode 39 of Talking Postgres, Adam Prout—distinguished engineer at Microsoft and a founding architect of Azure HorizonDB—joins Claire to trace his engineering journey from MemSQL to Postgres. We dig into shared-storage architecture and how HorizonDB pushes more work into the storage layer; why the team chose Rust; and what “good systems programming” looks like when being paranoid is a feature, not a bug. Along the way: startup vs big company tradeoffs, and how working on databases exposes you to so many interesting parts of computer science.  

Previously on Talking Postgres:
Talking Postgres Ep29: How I got started leading database teams with Shireesh Thota
Links mentioned in this episode:
Blog post: Announcing Azure HorizonDB, by Charles Feddersen & Affan Dar
CMUDB talk: HorizonDB: Co-Designing Postgres and Azure for Cloud-Native OLTP, by Adam Prout
Research paper: Socrates: The New SQL Server in the Cloud
Product page: Azure HorizonDB
Video of POSETTE 2025 talk: Scaling Postgres to the next level at OpenAI
Blog post: Scaling PostgreSQL to power 800 million ChatGPT users, by Bohan Zhang
Blog post: Supporting ChatGPT on PostgreSQL in Azure, by Affan Dar, Adam Prout, & Panagiotis Antonopoulos 
Docs: Azure Database for PostgreSQL
GitHub repo: pgrx
Discord: PostgreSQL Hacking server
Conference: PGConf.dev 2026
Conference Schedule: PGConf.dev 2026 Schedule

It’s rare for developers to genuinely love their database, so why does Postgres earn that kind of loyalty? In Episode 38 of Talking Postgres, Gwen Shapira, co‑founder of Nile, joins Claire to trace her path from operating Oracle at scale, through a long NoSQL chapter, to co‑founding a Postgres company that wasn’t originally meant to be one—after discovering how much Postgres quietly gets right. Gwen shares how she spun up on Postgres after years with other databases (shout-out to the Happiness Hints and a strong sense of curiosity), and why Postgres has made her appreciate Codd. We also touch on blogging as a career catalyst, the upcoming PGConf.dev conference (where a lot of PG20 work gets discussed, and where we’ll celebrate 30 years of Postgres)—and the #1 rule of consulting: it’s always the consultant’s fault.
 
Previously on Talking Postgres:
Talking Postgres Ep15: My Journey to Explaining Explain with Michael Christofides
Talking Postgres Ep24: Why mentor Postgres developers with Robert Haas
Talking Postgres Ep30: AI for data engineers with Simon Willison
Links mentioned in this episode:
Gwen’s company: Nile
Ardent Performance Computing blog: PostgreSQL Happiness Hints
Book: Just Use Postgres! by Denis Magda
GitHub repo: HypoPG
Wikipedia page: Codd’s 12 rules
Blog post: Transaction Isolation in Postgres, by Gwen Shapira
Podcast: Postgres.FM
Discord server: PostgreSQL Hacking 
Upcoming conferences & talks mentioned:
Conference: PGConf.dev 2026 on May 19-22 in Vancouver Canada
Conference Schedule: PGConf.dev 2026 schedule
CFP deadline: PGConf.dev Community Discussion Sessions deadline on 14 Apr 2026
Conference: PGConf EU 2026 in Valencia Spain
LinkedIn post: PGConf.dev 2026 conference t-shirt, celebrating 30 years of Postgres
POSETTE 2026 talk: The Rise of PostgreSQL as the Everything Database, by Varun Dhawan
PGConf.dev 2026 Panel: Real-Time Patch Idea Evaluation
PGConf.dev 2026 Roundtable: Unexpected Successes & Epic Failures by PostgreSQL committers
Calendar invite: LIVE recording of Ep39 of Talking Postgres to happen on Wed May 06, 2026