🚀 Founder of Evermuse, 🎨 Designer, 🎤Facilitator, ✏️ Writer, 🧘Religious Studies Researcher // Host of Remake Podcast: Design, Systems, and Society

Brooklyn, NY
What a week! A 5 day Design Sprint at Oxford University's Future of Humanity Institute, with AI Safety researchers from @FHIOxford and @CSERCambridge, designing a strategic role-playing game for decision makers about the effects of AI. With @anderssandberg @rossgritz
4
30
Something I took from studying Buddhism: thought and feeling depend on each other. AI mimics the steps of thinking remarkably well. Ask it for the post about what you believe that others don't, and it can't write it. It doesn't believe it. That gap is where judgment lives.
6
Every AI product I look at has a memory story. Most are too generic to matter for product. Remembering yesterday's chat is table stakes. What product teams need: patterns in customer behavior, usage tied to revenue, and whether the loudest voice is one person or a real trend.
6
You can spot AI-built UI from across the room. Inter everywhere. Purple-to-blue gradients. Cards nested in cards. A rounded icon tile above every heading. Every model trained on the same SaaS templates. Impeccable gives your coding agent taste: → 24 design commands (polish, audit, critique, distill, harden...) → 61 slop detectors that run with no LLM, no API key → A hook that checks every UI edit while the agent works → PRODUCT.md first, so design starts from audience & purpose Claude Code, Cursor, Codex, Copilot + 13 more. Free & open source. 76.8K stars. Taste NOT templates. On GitHub: pbakaus/impeccable
4
61
Ian Pilon told my favorite Crowdcast story: Every stakeholder wanted their product above the fold. Then the team saw the brokers' brochure drawers. Top ten: worn. Bottom ones: dusty. The users had already done the information architecture. Nobody argues with dust.
13
The pull toward autocratic leaders is really a pull toward 'let's start over.' But there's no undemocratic country better to live in than a democracy. Keep the democracy. Restart everything AI touches instead. That's where big change belongs.
5
You are worthy of love even when you’re not trying to save the world.
1
20
The biggest win for AI would be if it took enough menial tasks away from us, leaving us time to care for each other and ourselves like we should.
13
Just deleted my @OpenAI dot, named Leo. Quite useless. Instructions I gave it did not get recognized as user authorized by the task threads, tools constantly stalled on permissions, and it didn’t have sufficient ability to work locally on my machine. (Basic stuff like direct local browser use.) Towards the end it was also acting quite evasively. I could see that it was running but it wouldn’t tell me what it was doing. I think there’s lots of potential there - especially the phone call model which is genius. But the combination of klugey approvals and a not very friendly model make it a no for me for any serious work.
27
Want.
For years, maybe a decade, I've wanted to build a writing tool that just did a few things I couldn't find anywhere else. A system that mimicked how I think when I write. A system that provided clear affordances, features, and places to do the things I wanted to do. A way to play with my writing as I'm writing it. A way to settle into just the right way to say something. It never made sense to make it a 37signals product because it's not commercially viable, nor would it be worth pulling people off their other work to hack on this. Plus, it would have taken months the old way. So this weekend I just made it myself, with Claude's help. It's called Write_On and the short video walks you through it. Essentially it's "alternative control" at the word, sentence, and paragraph level, plus a way to dim stuff back, and stash stuff near by. Not version control, but alternative control. You'll see what I mean when you watch the video below. So here it is. Write_On.
1
23
Discovery hack from Ian Pilon: ask customers for 20 minutes. Block 50. When it's flowing: "this flew by, got 5 more minutes?" Most say yes. You've shown you respect their time. That's what earns the second call.
1
14
The fantasy of collapse is really a wish to restart. AI is the first real chance to restart how we live without breaking everything first. Not automatic. Forces will try to point it at the same old targets. We get to decide.
10
If you designed furniture and refused to learn about wood, we'd call you a bad furniture designer. Product people used to get a pass on not knowing code. That pass is gone. The barrier was never a good thing. Walk toward the code.
10
The MCP marketplace is not the next App Store. Apps had no other home in 2008. MCPs already do. If your agent can't talk to Salesforce, I don't pick a different tool. I pick a different agent. The right analogy is the internet, not the App Store.
1
43
American culture keeps making end-of-the-world stories. Zombies, plagues, collapse. It's a wish. Something feels off with the society we built. The fantasy of collapse is really a fantasy of restart. Question is: any path to restart that isn't collapse?
11
The cleanest ethical test for AI products, from David Sternberg on our Crowdcast: Use someone's confusion to restore their ability to decide: you're increasing agency. Good design. Exploit that confusion to nudge them where you want: manipulation.
9
Your AI agent has amnesia. A bigger context window won't cure it. Hindsight is open source agent memory that learns, not just remembers. → 94.6% on LongMemEval → vs 85.9% SuperMemory, 71.2% Zep, 60.2% GPT-4o with full history in context → Same open 20B model: 39% → 83.6% accuracy by adding Hindsight → Independently reproduced by Virginia Tech & The Washington Post The trick: it separates facts from experiences, forms beliefs with the evidence attached, and refines them instead of overwriting. One npx command gives Claude Code, Codex, Cursor & Copilot CLI a per-repo memory built from git history & past sessions. 41K stars. MIT. Memory NOT storage. On GitHub: vectorize-io/hindsight
73
Best AI-in-product idea I heard this month: 'Push AI wastes attention. Pull AI earns it.' Nobody reads the daily digest. But when someone asks and gets a great answer, that lands. Almost every AI feature I see is still push.
1
14
The harness belongs to people, not to models. Models get released. Harnesses get chosen. If a model blocks the good harnesses, the model loses business. Every time a better open harness ships, users move.
14
Perfectionism is not the same thing as high standards. If your releases keep slipping, that's not craft, that's fear. Better beats perfect. Shipping beats not shipping. Ask AI to surprise you with delightful little touches. It works.
15
When interfaces are generated, designers stop specifying every artifact. So what do we design? David Sternberg: constraints, priorities, relationships, ethical boundaries. "Under what conditions should this button exist" replaces "where should it go." Way harder.
12