I wrote a 6,000 words retro on my career so far. I found it useful to reflect on the last 40 years and fun to explore some interactions with chronology.
Time is unrelenting. Life is long.
https://t.co/iNJXoAyPuw https://t.co/I4XJE1NhFo
Watch on XSignals
Things I stumble on — sometimes interesting, sometimes relevant, sometimes just funny.
August 2026
Watch on XThe fastest way to become articulate on a topic is to write until you're no longer confused.
Introducing https://t.co/8MeR16MRQn, a tool to measure how well agents can read your site. Backed by @oradotai's research, you can run:
▪︎ Audits with 100+ checks
▪︎ Visualizations of agents using your site
▪︎ One-click prompts to fix problems
▪︎ A CLI for agents
/scandinavian-design
built a skill that turns any site into scandi minimalism!
npx skills add ericzakariasson/scandinavian-design
try it: https://t.co/8i3zHybIj8 https://t.co/wfPrk6338v
Watch on XOverall Preference Elo generally increases as Time to First Audio (TTFA) decreases, suggesting that responsiveness contributes to a preferred conversational experience. Gemini 3.1 Flash Live Preview - Minimal leads overall preference at 1,046 Elo with a 0.96 second TTFA, compared with GPT-Realtime-2 (High) at 914 Elo and 1.14 seconds, and Qwen Audio 3.0 Realtime Plus at 699 Elo and 1.54 seconds.
Foundation learns from agent sessions (Codex, Claude Code, Cursor, Slack, MCP) and maintains a self-improving wiki of team knowledge. Built on ChromaDB with versioning, lineage, access control, and SOC 2 Type 2 compliance.
AI ToolsDev Toolstrychroma.com · Aug 20
I’ve been looking forward to today for 3 years
Today we’re announcing Foundation - Chroma’s solution to memory
Our research preview of this technology builds self-improving memory from your agent sessions. Try it out at https://t.co/DpynS0vnSd https://t.co/IkTOBAD1tK
Watch on XDeveloper platform providing fast coding, build, and test environments across Mac, Windows, and Linux. Includes Devbox for AI coding agents and a high-performance drop-in replacement for GitHub Actions runners.
Dev ToolsAI Toolsnamespace.so · Aug 20
Serverless cloud platform for AI workloads. Provides Python SDK for defining compute environments, instant container boots, autoscaling from 0 to thousands of GPUs, support for inference, training, fine-tuning, and secure sandboxes for agents.
AI ToolsDev Toolsmodal.com · Aug 19
@kentdebruin I collected 70 different moats, and they're all still relevant
https://t.co/AnpR9P3bV6
Platform for building universal Android, iOS, and web apps from a single JavaScript/TypeScript codebase using React Native. Includes Expo Application Services (EAS) for cloud builds, over-the-air updates, and deployment.
Dev Toolsexpo.dev · Aug 17
Connected design canvas built on real HTML/CSS for teams shipping with AI agents. Design equals code with no translation layer. Desktop app and MCP server let agents read/write the canvas, sync tokens and components with codebases, and work with real data. Raised $34M Series A from Accel and ICONIQ.
Dev ToolsAI Toolspaper.design · Aug 16
every time you intervene and correct your agent, you should think about how to eliminate it entirely.
in order of value:
1. categorically eliminate the problem through better architecture or choice of data structures
2. turn it into a lint rule or test so CI catches it
3. turn it into a skill or rule
4. have humans review the code to catch it (ngmi)
here's how I've been using Grok Bot https://t.co/JFt0eTz37O
Aug 15
Aug 15
Mobile development platform that helps teams build, verify, and ship apps with confidence. Provides cloud iOS/Android devices, AI agents for runtime testing and verification against the live app, integrations with Cursor and CI, and a living Atlas source of truth for product context.
AI ToolsDev Toolsrevyl.com · Aug 14
Video is up!
https://t.co/LCp00IRbK7
New: A map of the most important skills in AI Engineering. https://t.co/VVkn1Dqp1N
Today we're sharing the Notion Knowledge Board.
We get to observe a wide range of real-world AI tasks at @NotionHQ: meeting follow-ups, support tickets, sales pipeline updates, market research, all the good stuff. Knowledge board is a live, ongoing eval of how models handle real knowledge work measured on live anonymized traffic and scored by an ensemble judge we built in collaboration with researchers at Anthropic and OpenAI.
The takeaway is the tradeoffs: cost per task variance is high, and some of the fastest models are also very capable. The right model genuinely depends on your work.
We will update as new models ship: https://t.co/BMmwQV9W6m
Search and filters always cover the full archive.