Jan Ole Suhr (@janole) header
Jan Ole Suhr (@janole) profile picture

Hands-on, open-minded Product Engineer since 1986. Entrepreneur at heart. Former InsurTech CTO/CPO. Creator of Gravity, the social net app for Nokia w/ 1M users

Berlin, Germany
Joined July 2007
Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)Photo by Jan Ole Suhr (@janole)
GLM 5.2 has taken control of a tmux session with Claude Code in it. It's watching and waiting for approval prompts. Then it's trying to approve or reject them, but that tool call itself is running through another GLM 5.2 approval agent 🤯 The fact that I can tweet this means neither of those agents accidentally wiped my MacBook yet ... 😂🤣
3
351
Always good to analyze all your sessions. Apparently, Opus 5.5 was switching away from using Bash for reads/writes/edits from 11-05 to 11-07. Back to normal now. I didn't notice any difference in quality, and usage was feeling normal, too ... 🤷
1
3
127
Argh! I seem to have been targeted by another Anthropic A/B test. Probably worth monitoring regularly. It's when you see lots of Bash usage ("sed ...", "cat ...", ...) instead of Read/Edit/Write. For me, "bashFirst" works better.
2
184
It's nice that you can just use Claude to analyze itself: "Since about 09:13 UTC today the server has Bash-first switched on for us anyway, so nothing urgent is wrong right now. ..."
56
Jan Ole Suhr retweeted
Cybersecurity expert #1: AI is becoming superhuman at finding vulnerabilities and soon no computer will be safe Cybersecurity expert #2: also our private key is '4' and if you mash the keyboard long enough it logs you in as root
*sshole is out if you run sshd on alpine, openwrt or any musl box with password auth on: update now official writeup: sshhole.com
5
37
3
244
5,824
Wild bug …
Background subagents in Claude Code now use less of your usage limit. A running subagent updates its status line every 30 seconds, and each update used to re-read its whole conversation. Now Haiku 5.5 writes it from short notes instead, so each update costs about 99% less.
1
163
When the agent starts to talk to sub-agents although it shouldn't and I'm reading ... "Great work — the operator loves the Plan tab." ... I'm thinking to myself: are they alive!? 🥰
4
219
Jan Ole Suhr retweeted
Sorry bro, weekend is for the lazy, I let my robo clean up my desktop (sound on)
120
173
58
3,045
268,732
Jan Ole Suhr retweeted
the moat is to be a solo founder
there is no moat
4
1
44
1,958
🧐 „Claude Code burns ~15–20% of your tokens on a 3-word status line. Here's how to turn it off“ gist.github.com/flamerix-ai/… Not sure if that’s really true. Can’t reproduce it (I am using Claude in my own harness with my own remote-/sub-agents only.)
155
Testing the harness ... Opus 5.5 driving 40 sub-agents of deepseek-v4.1-flash-1017. Cache hit rate looks healthy, I'd say 🥰
2
2
186
Jan Ole Suhr retweeted
feels like there needs to be a new way to distribute opensource software such that people can modify via agents but also get upstream automatically merged in coding agents can do this but they won’t do it automatically
experimenting with a lil personal project: an AI-native macOS dock with custom modes (writing, building, etc.) to help me stay focused my agent is able to customize mode behavior and add new modes, it uses jev to classify my activity + notifications in real time & act on them
297
83
46
3,476
213,295
I think I must be in some test group with my Claude subscription ... Third day into the "Claude week" and I have a full day of usage (~14% points) leeway ... 😍🥳🎉
2
2
253
Finally started to use Tailscale back in August before going on vacation. I can only suggest to try it out, and add some instructions / scripts to your AGENTS.md to automatically show the corresponding Tailscale URL alongside every local URL.
1
196
Jan Ole Suhr retweeted
I tried to use and like Mastodon but didn’t last (too many unwritten social expectations on what I acceptable to post & how) and I self-censor posting about AI on Bluesky because I’m tired of the anti-AI responses there from devs. Today, X is the most positive forum on tech
The poll so far between Twitter, Bluesky and Mastodon. I think it looks less absurd than last time, but still pretty clear that the base is divided. xitter.gq/mitsuhiko/status/21143…
77
19
8
664
40,512
The feeling when "you" have just implemented "your" own React DataTable component with virtualization ... and the weekly usage of your Claude subscription is still at 7% on the second day of the "Claude week" ... 😍🥰💪
1
1
190
Who's running their software factory from the wrist? (I am ... 💪 ... it's not my arm, though, I wear my Apple Watch on the right hand 😅)
🤖 Made with AI
3
5
441
... and usually, I don't approve "bash" right away but click on the next button that's barely visible: "Hand over to agent" 🤠
70
Jan Ole Suhr retweeted
DeepSeek Status 1.4 released. New display support for LithosAI balance. GitHub: github.com/owenzhao/DeepSeek…
1
1
2
98
The software factory is improving ... 🥰
1
3
234
I just measured what it "costs" my software factory to ship one work item, from first prompt to merged PR, review included. Opus 5.5 writing the code, Fable 5.1 reviewing it, roughly priced at API list rates: Median $5.05 per PR. 166 PRs in 14 days.
3
8
298