Recent posts
A little project today: a custom app to teach me more colors. View, match, and name the 200 most popular color names (in English). $0.40 with GPT-5.6 Luna. Single page app that loads instantly - give it a try. https://abramjackson.com/p/colors/
Most important feature to copy from LinkedIn: by-lines visible below every name on every comment. I need to see everyone's grossly overstated titles!
Does talking about Copilot and AI agents leave you confused? It doesn't have to! Learn everything you need to build agents and accelerate what you can do with Microsoft 365 Copilot in the Agent's Playbook webinar beginning next week. Register now! https://developer.microsoft.com/en-us/reactor/series/s-1696/
All this time, and I've never learned who lives in a pineapple under the sea.
Challenge level impossible: come up with a business idea that every chatbot rates 1/10
You all weren't doing this? I recommend getting on the github Copilot train. My agents are constantly spinning up more agents in separate repos.
New in Claude Code: your sessions can now message each other. Instead of having to re-explain yourself in another session, you can now tell Claude to do it. It sends a summary (not your history or files), and the other session picks it up mid-task.
time spent discussing OKR definitions is inversely correlated with achieving Objectives
Reply bot that semantically searches memes and customizes them to respond to your tweets
a mad lad over here posting this chart
On EmulatorBench, it built SEGA Genesis and Game Boy Color emulators from scratch in Rust, reproducing target hardware behavior against diagnostic tests.
the existence of nonsofic groups has me wondering what else is out there...
a nonsofic group is an infinite system whose behavior cannot be approximated by finite models Asimov's Foundation series was all about modeling large systems. the Mule in Second Foundation shows what happens when the group can't be approximated
Don't forget: the coding agent takeoff started with Gemini 3.1 Pro. Let's see what they're cooking.
Let's get a "Ask Grok" button there too.
Microsoft-owned LinkedIn just added a “seems like AI slop” button to report posts that don’t feel human-written. It’s part of a broader push to get rid of AI slop on LinkedIn https://www.theverge.com/ai-artificial-intelligence/973384/linkedin-seems-like-ai-slop-button
If you purchased this book without a cover you should be aware that this book has been destroyed by Anthropic to feed Claude. It was reported as ‘unsold and destroyed’ to the publisher and neither the author nor the publisher has received any payment for this ‘trained’ book.
I am merging infinity percent more PRs, because as a PM I wasn't doing it before.
During Microsoft’s rollout, engineers merged 24.9% more PRs in weeks they used Copilot CLI than in their own no-use weeks. And the lift increased with usage, reaching 50.1% in weeks with 5+ active days.
I'm using Whisper and an agent to listen to everything I say, then it edits Wikipedia to make it true
How can we make this happen, as quickly as possible?
Open-weight models are essential to a healthy AI ecosystem. Together with others across our industry, we are outlining a path for open-weight models to strengthen American competitiveness and expand economic opportunity, while protecting national security. https://www.microsoft.com/en-us/corporate-responsibility/topics/open-weight/
