The Cathedral, the Bazaar, and the Winchester Mystery House: an attempt to put our era of sprawling, idiosyncratic tooling in context. www.dbreunig.com/2026/03/26/w...

Bryce 1.0 (1994)

At the DSPy meetup a couple weeks ago, Kshetrajna Raghavan walked through how DSPy helps with efficiency, modularity, and performance at Shopify scale: www.youtube.com/watch?v=bxTo...

I don't think I've ever felt more like the meme of the sweaty dude agonizing over what button to press.

I used Google Stitch, the AI frontend design tool whose announcement had people panic selling Figma. I am beginning to think people should try these tools before they react...

2016: Companies argue they’re a software company to get higher valuations. 2026: Companies argue they’re not a software company to get higher valuations. www.bloomberg.com/news/article...

If you’re buying the first campaigns on a platform, you shouldn’t be expecting data saying the ads “worked”. The point is being first, learning, getting the awards, and being seen as a leader.

Based on my friends in the military, it makes total sense that the two most regular leakers of military info are Warthunder forums and Strava. www.lemonde.fr/en/internati...

“This is, I’m sorry to say, altogether goofy, but nonetheless beautifully expresses a desperate romanticism that seems endemic to the United States: a desire for roots in some place other than this, our graceless national project.”

Two years ago I build a dataset from Wikidata for exactly this and am so happy someone else built it!

“Don’t go in the kitchen. The robot’s making dinner on a Ralph loop.”

Apple's AI leverage is their developer tools. If the next iOS adds a feature where an AI can inspect and parse the programmatic layout of an app (either when onscreen or off), without doing screen-grabs and image processing, no one will complain that they're licensing Gemini.

🫠🫠🫠

This is THE issue. And honestly, picking another provider doesn't solve it. I fully expect the military to eventually train their own model. www.dbreunig.com/2026/03/01/t...

*monkey paw curls* a niche obsession with training run variance is sure gonna become important soon ...

A DSPy tutorial: learn the fundamentals while building a deep research agent. Covers Signatures & Modules: what they are and why they matter. No DSPy experience required. www.cmpnd.ai/blog/learn-d...

The inaugural ACM CAIS 2026 conference on AI systems is soliciting additional reviewers. If you are interested in reviewing, please fill out this short form. Reviewers should expect a load of 4-5 papers, with reviews due April 3, 2026. docs.google.com/forms/d/e/1F...

llms.txt is a failed standard. Supply-side standards (like Do Not Track) don't work; creating demand is the only lever connected to anything. dri.es/markdown-llm...

Yesterday I gave a talk sharing what I learned building a no-code library, why spec-driven development is a feedback loop not a straight line, historical parallels for our current moment, and shared a PoC tool for keeping specs/tests/code in sync. www.dbreunig.com/2026/03/04/t...

🤔

The Anthropic/DoW conversation centered around terms of use, but the real issue is embedded judgment. www.dbreunig.com/2026/03/01/t...

Who is this for?

Two beliefs about the agentic coding narrative: 1. Most devs underestimate what they bring to the agent. Agents amplify skills. 2. Most coding agent written software we see are personal tools, not capital-P products. Shipping remains hard. www.dbreunig.com/2026/02/25/t...

Time for another SF DSPy meetup. This time focusing on DSPy in production use cases and RLMs. Engineers from Dropbox and Shopify will be sharing case studies and answering questions, and Isaac Miller will be walk through dspy.RLM. Come! luma.com/je6ewmkx

Please, someone: build an API designed for agents and claws that answers geo questions, using Overture and OSM data. Google Maps pricing tiers are not made for this and designing an API that translates LLM logic into geo queries is not insignificant. Also, OSM is getting unsustainably hammered.

This is so absurd. If you want to make a bet on agentic coding crushing software go for it – but don't _start_ with things still running on COBAL. (Also, IBM has been selling agentic COBAL stuff since...2023?)

Here's a fun Claude demo: A friend and I are reading "The Worst Journey in the World", a memoir about a 1910 trip to the South Pole. We couldn't grok their route from UK to Cape Evans. I dropped the full text in Claude and in 7 turns had a map: claude.ai/public/artif...

This post is getting my favorite type of response: no matter someone's position, people are using it as evidence that proves their point. www.dbreunig.com/2026/02/21/w...

The state of coding agents can be summed up by this anecdote: Anthropic spent $20k on an agent swarm implementing (kinda) a C-compiler in Rust, but desktop Claude is an Electron app.

Cold evening in St Helena last night.

Love when The Economist just wallows in it.

Great new API for the GEPA prompt optimizer: optimize-anything. One of the immediate use cases for optimize-everything is this demo that autogenerates Skills, given a repo: gepa-ai.github.io/gepa/blog/20...

🤔

The deadline for paper submissions for the ACM Conference on AI and Agentic Systems (CAIS) is approaching, February 27th. If you're working in this domain, you should apply. It has a killer committee and a partnership with The AI Engineer World's Fair. www.caisconf.org

My not-so-monthly newsletter just went out, covering a book on context engineering I'm editing for O'Reilly, upcoming events you should checkout/attend/submit a paper to, and recent writing. buttondown.com/dbreunig/arc...

All the content created by clawdbots isn’t passing the test I keep applying to image, music, and video gen tools: if this wasn’t created by an AI would it be novel?

Srihari Sriraman and I took a deep dive into coding agent system prompts to understand their structure, similarities, and differences. www.dbreunig.com/2026/02/10/s...

Two Harbors Catalina 1, LosAngeles County, CA 🗺33.4413, -118.4861 🧭264° ⛰354 ft https://ops.alertcalifornia.org/cam-console/2468

Why I’m excited about RLMs www.dbreunig.com/2026/02/09/t...

12 yeas ago

No

A napkin drawing of what you want AI to make is virtually guaranteed to be more interesting and I wish more people would try it.

RLMs are an incredible strategy for mitigating context rot. DSPy is the easiest way to use RLMs. Check out this overview: www.cmpnd.ai/blog/rlms-in...

A whenwords update, looking at more examples of spec driven development, including just-bash, monty, and Anthropic's c-compiler. www.dbreunig.com/2026/02/06/t...

using claude code at a standing desk feels like DJing

If you're doing applied AI research (especially system design, benchmarks, evals, efficiency, or ops) you should be submitting to the Conference on AI and Agentic Systems... caisconf.org/pages/cfp/

I guess we’re gonna end up building a coalition around the constitution. Full circle.

Anthropic comms is crazy. “Here’s a 38-page essay about the existential threat of AI and how underprepared we are.” *30 min later* “Also you can now connect your B2B SaaS apps to Claude.”