*monkey paw curls* a niche obsession with training run variance is sure gonna become important soon ...
10.03.2026 16:43
👍 67
🔁 3
💬 2
📌 2
*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...
That said, I extracted the code I use on my Jekyll site that provides markdown versions of posts and published it as a Gem: github.com/dbreunig/jek...
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...
Lots of historical precedents here: overwhelming amounts of code birthed software engineering.
We’re figure out a framework that works, it just won’t be “fun” for those that enjoy Gastown and similar.
Just wrote/spoke about this: www.dbreunig.com/2026/03/04/t...
Every jersey has a different personal injury lawyer sponsor
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...
🤔
Same! Still friends with many.
The adults that got added to the room had some of the most impressive CVs I've ever seen...and were the exact wrong people to guide it at that time.
I interviewed at Tumblr very shortly after the adults had been added to the room and was asked what the biggest problem with Tumblr was. My answer was, "Success on Tumblr means leaving Tumblr." They should have been Patreon, Substack, etc etc.
The Anthropic/DoW conversation centered around terms of use, but the real issue is embedded judgment. www.dbreunig.com/2026/03/01/t...
But they gotta play the weird card: we don't support autonomous lethal use........now.
Who is this for?
Theres so much great about that cover and album.
My favorite was reading Pharrell saying his #1 inspiration for his performance was the lead singer from America.
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
I don't see the problem to fix.
Yeah! I bet it's lots of restaurant requests and place searches within town/city/neighborhood parameters or "near" something.
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?)
Talked to a few people this weekend who wondered, "why are LLMs writing in Python? it's not a language for LLMs?"
Buddy, at some point humans gotta read it.
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...
I think the Boris update is notable because a) it doesn't address the central complaint (it's a slow app!) and b) if you click through to HN he is getting dragged for making big claims just last week about code being solved.
Familiarity is so nice (especially when there's strong conventions) for debugging...which is when you REALLY look at the code.
Funnily enough, I went through a multi day technical interview at Facebook circa 2012. I think the reason I didn’t pass was one session with their head of mobile dev where I argued native iOS was the way to go.
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.