Belgian. Building EU products. Currently working on Vist (https://xitter.cf/t.co/jSSKBCNHk2)
Belgium
Joined January 2026
- Tweets339
- Following191
- Followers32
- Likes375
Worth listening to all the samples on their site. O-M-G.
We believe voice is the most natural way to interact with intelligence, so we created Dessa, the most authentic full-duplex speech model ever built. Talking with it feels like talking with a person…full of emotion, nuance, pauses, and a natural cadence.
block.xyz/dessa
Ooooh yeah
I have 🥁🥁🥁🥁
30!!
whole followers now.
Almost there! How many more until total world domination?
MCP tip: your tool descriptions are prompts. Every word counts.
Vist's update_task replaces the whole steps list, so the description says: "call get_task first to avoid dropping steps."
One sentence. Agents read it. And if you don't catch them on nerf day, they get it right!
Belgian Product Designer by day, building Vist by night: notes, tasks & memory that my agents share over MCP and CLI.
Looking for people who run more than one agent on a project, and are not in a huge org. How do you keep agents in sync?
Reply with your setup, I’ll read ‘m all
Yesterday's Vist agent loop run:
- notes now multi-select, bulk move and delete
- CI is green again... the system specs never built Tailwind. lol.
Conclusion: one feature (or UX gap close) that had been bugging me for weeks, and one chore I'd been ignoring for days. Fair!
Fixed a fun one in Vist last night: if you typed "due friday" on Saturday... the task could land on the Friday before. Overdue on arrival. 😅
Relative dates only go forward now. Type an actual past date and quick-add tells you instead of quietly filing it.
Vist tasks have steps now.
Not subtasks - no, just a title and a checkbox, so an agent can break a task down over MCP and tick things off as it works, and I can see how far it got without asking.
What if a step needs its own deadline, you say?
Well... then it's a task. Sorry!
Switched to an agent loop for my Vist app development and it is amazing how faster I can move.
I really was the bottleneck. Damn.
Thanks Grok @bot for pushing me to add the agent kanban board and /vist-next command to Vist.
David @ vist retweeted
Head of Design at Stripe on how to stop creating zombie UIs
If we only use AI to make the things that we already make, just faster, then we're definitely missing out on the most interesting part.
From my talk at @lennysan’s Summit, here are four ideas for raising the ceiling, not just the floor:
- Have a point of view [5:38]
- Encode your standards into the machine [9:07]
- Don't confuse done with good [12:21]
- Unleash creativity and artistry [16:19]
Full talk here: youtu.be/GLvFTMtw4Jk?si=-5Ts…
David @ vist retweeted
Going thru all responses. Such wild variance!!
General trends:
- PRs per eng is up
- Spend is anywwhere from $100/dev/month to ~$5,000. A few outliers in Big Tech at $20K+
- Very very very few teams ask about the ROI on the spend!! Or know how to measure. More want to tho
David @ vist retweeted
jev new Box primitive is Yolo for everything
your favorite mug, your cat, your specific plant, it can take all that and more as inputs and find it
Friday morning ritual:
Coffee, then a couple of overnight agent PRs waiting for me.
Some I open carefully. Some I just glance and merge.
Progress?
David @ vist retweeted
UI agents like Claude Design are about to split companies into two groups:
1. Companies that know you design the experience. They'll give experienced designers superpowers with these tools.
2. Companies that think design means interface. They'll cut as much design and as many designers as they possibly can.
Which one are you?
Absolutely agree. I switched to OpenCode and now I use it through T3 Code. Connected Vist so whatever model or harness I use, it always has all my knowledge and project state.
Total.
Freedom.
No lock-in! Not tied to US oligarchs either.
I'm convinced that relying on a frontier coding agent (Claude Code, Codex, ...) is a mistake, for engineers and their companies. As good as they are, as much of a Claude Code die-hard as I've been, I no longer recommend them.
The "best frontier model" is a moving target: it went from Opus to Fable to Sol to Opus again in 4 months.
These agents are built for their lab's models and plans. Yes, you can run other models, but it's not native or in their interest.
You may be locking yourself into a worse-model future that's hard to leave. As with IDEs, there's a lot of inertia in tools we use all day, and in the skills, prompts, MCP servers and hooks we add to them.
It's like Formula 1: engine suppliers come and go, the teams stay and keep improving the car.
Instead, get comfortable with @opencode, @t3dotcodes or Vercel's fx. Configure them well, use a gateway like Vercel's AI Gateway or OpenRouter, and switch to the best model the day it's out.
PS: I work at Vercel, but this isn't a Vercel point of view or a criticism of AI companies. Closed models are perfectly fine. Tying your tools to one lab is a dangerous bet.
1/ Vist 2.0 is out.
I gave my coding agent a kanban board. It picks up tasks I approve, opens PRs, and reports back. I just merge.
Built from plain notes, tasks and labels, over MCP.
3/ Favorite moment: on the last PR, the agent left one box unticked and wrote why. The plan asked for something that wasn't possible yet.
I'd much rather have that than a green checkmark that lies.
4/ Free plan, EU-hosted, works with Claude Code, Cursor, Le Chat or anything that speaks MCP!
How it's wired: usevist.dev/blog/agent-kanba…