Hey folks. I’m researching how people are extending their agents. I would appreciate if you could fill out and forward this survey.
How to find and use agent extensions
forms.gle/ENgpYg9ywqLv...
I saw that with pi.dev was surprised at first but then it made so much sense. I love it on so many ways.
Yesterday I had the opportunity to present my ideas how Coding Agents can be used for non-coding cases at the Cologne's Vibe Coding meetup. I've shown my CRM setup with Claude Code. And we have had some interesting back and forth on terminals and markdown mess. ;-)
● You're absolutely right!
The interesting part when working agents is memory management. Users can and should actively manage memory. But that is very different from what we are used to. Now we can start from scratch. A powerful concept. Especially if you combine it with previously taken snapshots.
Best description I’ve seen so far:
Skills are lazy loaded prompt engineering.
news.ycombinator.com/item?id=4625...
Who is building agents with Vercels AI SDK and is up for a chat?
My app felt kind of slow until I realized that it hibernates and that the VM boots in 1.2 seconds. @fly.io #firecracker
Anyone have some good UK-TwoStep/Garage music recommendations?
I’m battling daily on giving up control yes or no. Last week I gave Claude’s agent SDK a try. The amount of filesystem tool calls made me crazy.
Here is another take on 12 factor but especially on agents github.com/humanlayer/1...
Nice. Never thought about it as full fledged PaaS replacement. Interesting thought experiment: What would Anthropic or OpenAI need todo to host the apps we’ve build on traditional PaaS?
There are basically only two positions in the debate about AI.
1. I’ve barely invested any time in learning how to use it effectively. AI sucks.
2. I’ve invested in learning how to use this tool. Holy cow, it’s transformational.
Position #1 has lower barriers to entry.
Yesterday, we witnessed the App-Store moment of AI. The AppSDK allows developers to embed full-fledged Apps directly into ChatGPT. If you combine the transformation of the iPhone App Store with adoption pace of ChatGPT, the implications are massive: www.linkedin.com/feed/update/...
A friend asked me: “What do you think about the future of project management?” Here is my hunch:
Tomorrow’s project management will connect requirements, decisions, and code in a single traceable chain — every “why” and “how” and change is transparent to everyone.
www.linkedin.com/pulse/how-sp...
Thanks for stopping by. #cypresshill #bonn
The confidence is a function of the ignorance.
Here's video, slides and a detailed annotated transcript from my talk at this week's AI Engineer World's Fair conference in San Francisco - "The last year six months in LLMs, illustrated by pelicans on bicycles" simonwillison.net/2025/Jun/6/s...
Am Freitag gibt es Demos.
The recording of last week's meetup with Jason Liu, David Soria Parra (Anthropic), OpenAI and @sa.muelcolvin.com is now live on youtube.
Notably this includes David's plan for the future of MCP.
youtu.be/GqufkaLJV44
I've hacked together an MCP server to query all #KubeCon videos or any other Youtube Playlist: github.com/luebken/play...
e.g.
claude.ai/share/3fbee0...
e.g.
$ youtube_transcript_api bTVavLFH7XQJMN9D9UVZn4 --format text | llm -s "These are the captions of a presentation at KubeCon 2025. Please summarize it."
Michael Krenshaw, a staff software engineer at Intuit and an Argo CD lead, shared the progress and future of Argo CD at KubeCon 2025.
Good morning #kubecon. The CNCF has already started posting www.youtube.com/@cncf/videos of the presentations.
To summarize and gauge whether I want to see a presentation, I use:
$ pip install youtube-transcript-api
$ pip install llm
$ llm keys set gemini <your_api_key>
Great post by the Zalando engineering on how they used LLMs for migration of UI component libraries. Bonus: Using my favorite tools continue.dev and llm. engineering.zalando.com/posts/2025/0...
But who is Elizabeth?
wish our’s were on time.
