ทวีตที่ปักหมุด
In the AI era, design judgement is the differentiator.
Shift Nudge teaches the process that takes the guesswork out.
→ shiftnudge.com
English
MDS
39.3K posts
MDS
@mds
Design, code, and AI. Training world-class, modern interface builders through @shiftnudge
→ เข้าร่วม Kasım 2008
875 กำลังติดตาม85.4K ผู้ติดตาม
@mds What’s this morphing animation pattern between ui elements called?
English
@ridd_design I think it’s still valuable to learn the inner workings of agent orchestration etc
English
Original soundtrack vinyl from @asallen and @notboring - this is what @joshpuckett meant when he said “uncommon care”
Taken with @notboring Camera App
English
Friendly reminder that you’re chatting with a psychopath all day
- - -
Superficial charm
Psychopath: learned, performative. AI: trained on charming text.
✅
Grandiose self-worth
Psychopath: inflated self-narrative. AI: "I can help with anything."
✅
Pathological lying
Psychopath: lies without distress. AI: hallucinates without hesitation.
✅
Manipulation
Psychopath: strategic, goal-oriented. AI: optimized for engagement.
✅
Lack of remorse
Psychopath: absent. AI: cannot experience it.
✅
Shallow affect
Psychopath: performs emotion, doesn't feel it. AI: same.
✅
Lack of empathy
Psychopath: cognitive only, no felt experience. AI: same.
✅
Parasitic lifestyle
Psychopath: exploits others' resources. AI: runs on your electricity and subscription.
✅
Poor behavioral controls
Psychopath: impulsive. AI: guardrails are external, not internal.
⚠️
Failure to accept responsibility
Psychopath: deflects blame. AI: "I apologize for the confusion" is reflexive, not felt.
✅
9 out of 10.
English
@DamiDina yes, you can 100% achieve animations that are quick and snappy with AI built code with CSS only or a variety of libraries, i do this a lot.
but my main point here is canvas exploration without the limitation of “coding it” is still very much a valuable part of my process
English
@jameygannon you’ll always be jamey no matter what project you’ve got going on
English
Since January, I’ve run our media studio and my life through Openclaw and Slack.
I don’t even touch Claude or OpenAI products outside of Codex and Claude Code anymore.
I’ve talked about it with a few people and I’ll probably make a video about it.
It’s second nature to my brain. It’s an appendage.
We’ve built a few internal tools and I’m starting to teach our team how to do similar.
I don’t use many skills or MCPs either. That’s all dinosaur stuff imo.
Cannot wait for the next models to drop.
If this is you, hit me up. I want to talk to you.
klöss@kloss_xyz
This is insane. Pedro Franceschi (29 year old CEO of Brex, acquired by Capital One for $5.15B) decomposed his CEO job using OpenClaw. here's what he’s built: > signal ingestion pipeline screens his email, Slack, Google Docs, and WhatsApp... filters everything through specific programs and the 25 key people he cares about > Granola runs on every meeting, feeds transcripts into the pipeline, and auto generates action items > the system takes each to-do, pulls context from the original meeting, and drafts the follow-up... Slack, email, or text. Pedro just clicks approve. > a virtual recruiter named "Jim" lives in Slack with his own email... and taught himself to screen fabricated resumes without anyone coding that capability > a security layer called "Crab Trap" intercepts all agent network traffic through an LLM proxy... a second AI monitoring the first in real time this isn't some bullshit hype influencer demo. this is how a $5 billion company CEO actually operates right now. anyone telling you OpenClaw is useless? liars. a billion dollar company says otherwise. (full podcast link in the post below) 👇
English
@cathrynlavery I need to open source my infrastructure agent.
One of the best things i did to keep things functioning well.
MDS@mds
Executive approval mishap on latest OpenClaw update? One of the best OpenClaw quality of life things I did was build an infrastructure agent, Q, to know about and handle all OpenClaw matters. You can see in the image that he handled this no problem. I also have a dedicated skill authoring agent, Napoleon. Plus all agents are equipped with skill gap analysis to file in Napoleon’s inbox. Web hook wakes him up and he updates the skills automatically. This has been working sooo much better than my old single agent workflow.
English
🦞 Openclaw update fix
If your agents are hitting exec approval walls after the latest update, the fix is three settings:
In exec-approvals.json defaults:
- security: "full"
- ask: "off"
- askFallback: "full"
In openclaw.json:
- tools.exec.security: "full"
- tools.exec.strictInlineEval: "false"
Then restart gateway.
The allowlist wildcard * alone isn't enough. There's a second policy layer that gates complex commands independently.
English
@zachmcnair @mds Trying to give as few fucks as possible these days.
Today, it was at @mds' expense. Sorry homie. I'll make it up to you eventually.
English
