design, AI, web dev, data viz, tools for thought ✨ @shv, previously R&D @GitHubNext, design @AdeptAILabs
🍃 Oakland
- Tweets 6,592
- Following 5,081
- Followers 36,218
- Likes 14,530
Amelia Wattenberger 🪷 retweeted
new in Intent today:
- Home: all your projects in one place
- list & board views
- search, filter & pin workspaces
- review GitHub PRs with inline diffs
- start workspaces from Linear issues & PRs
- diagrams render in Preview
- export diagrams as SVG
- simpler command palette
- smarter settings search
1
3
28
6,596
Amelia Wattenberger 🪷 retweeted
and you can get it now....
steer agents across devices, review their work in browsers, before & after screen shots, diagrams, code changes, etc. run terminals, scripts, custom agents across providers. all from your phone.
intentapp.dev
designers, you need a gorgeous mobile app for managing lots of agents across lots of devices to crush important client requests like...
1
5
27
7,725
Amelia Wattenberger 🪷 retweeted
designers, you need a gorgeous mobile app for managing lots of agents across lots of devices to crush important client requests like...
4
2
75
17,779
Amelia Wattenberger 🪷 retweeted
I'm looking for a kick-ass product person who wants to do two things:
1. Get deep into the operations side of early-stage startups.
2. Completely rethink HCI.
You don't have to be a traditional PM, but you do have to be:
• Product-obsessed.
• Design- and/or engineering-adjacent.
• AI-pilled (though it's OK to hate the term "AI-pilled").
• In person every day in Palo Alto. (We have a beautiful office, great food, and an incredible, talent-dense culture.)
Reve's next chapter is going to be big, and we're moving fast. DM me to learn more. (No contractors, please.)
23
6
134
11,090
we need ways to make it clear what is…
- explicitly specified
- hinted at
- an agent assumption
any amount of the above is fine, but once code is in the codebase there’s no way to tease out what’s what
we've mostly gotten past the slop issue in this new world of working together with AI
the next problem is people keep accidentally undoing functionality
this is happening because everyone is more in everyone else's business. hit an issue? fix it right away without nagging anyone
the problem is it's unclear how intentional the current behavior is and what absolutely should not be rolled back
you can say tests should be this. but they don't capture the fact that you discarded seemingly valid solutions 1-4 and settled on 5. someone later (an agent) might thing "oh 3 is simpler"
any additional document also doesn't feel like it solves the problem because it's yet another thing that can drift
where is the source of truth for how your software is supposed to work?
5
2
36
3,804
how do we escape the chat trap? Maybe intelligent interfaces
wattenberger.com/thoughts/ai…
10
5
217
15,595
Amelia Wattenberger 🪷 retweeted
1. visual reviews
agents take before & after screenshots of their UI changes and place them side by side letting you judge results without reading code.
you manage agents, they manage more agents, those agents manage even more agents. chat threads break with that scale.
new article on useful views for what agents are planning, what they've already done, and how best to steer them: lukew.com/ff/2165/many-agent…
5
58
9,199
Amelia Wattenberger 🪷 retweeted
This is the future. As “intelligence” (read: ability to turn inputs into outputs) becomes a commodity, the interface is where you can create most value to the user.
we're back at the command line!
there's so much good thinking to mine from from the last time we transitions from terminal -> gui. had some fun writing down a few thoughts
would love to hear what you think
wattenberger.com/thoughts/ai…
1
2
858
Amelia Wattenberger 🪷 retweeted
Excellent piece from Amelia Watternberger: AI is Still in Its Terminal Era.
we're back at the command line!
there's so much good thinking to mine from from the last time we transitions from terminal -> gui. had some fun writing down a few thoughts
would love to hear what you think
wattenberger.com/thoughts/ai…
2
24
2,287
Amelia Wattenberger 🪷 retweeted
This is my fav way to code with agents right now. Really well built tool.
the broader point here: you manage agents, they manage agents, those manage more agents... chat threads break with that scale.
in this demo:
- a visual review of changes agents made
- diagram of how the system works
- arrange those views & more however you want
1
5
1,416
Amelia Wattenberger 🪷 retweeted
many ways to stay on top of many agents:
new stacks UI in Intent makes seeing/knowing/steering large amounts of agents even easier.
3
2
56
15,666
Amelia Wattenberger 🪷 retweeted
Put together a page collecting all the tabletop demos so far:
tabletop.grantcuster.com
4
21
1,803
really enjoying how these interactions turned out!
I put off making these examples until the last minute, but they're what finally convinced me to share. it's a just a small thing, but encourages the thinking that we can have tactile, more direct interfaces while leveraging our AI overlords
we're back at the command line!
there's so much good thinking to mine from from the last time we transitions from terminal -> gui. had some fun writing down a few thoughts
would love to hear what you think
wattenberger.com/thoughts/ai…
4
77
3,607
the potential to create entirely custom environments for creating is really only held back by our imagination
it took one prompt to add an editing surface directly on the post itself. "let me edit the text and symbols with a keybinding and save it to disk"
we're back at the command line!
there's so much good thinking to mine from from the last time we transitions from terminal -> gui. had some fun writing down a few thoughts
would love to hear what you think
wattenberger.com/thoughts/ai…
2
3
58
3,733
Amelia Wattenberger 🪷 retweeted
everybody moved from Terminal UI to combing coding tools with chat clients, but we don’t think dev tools have reached their final form yet.
8
4
72
11,932
we're back at the command line!
there's so much good thinking to mine from from the last time we transitions from terminal -> gui. had some fun writing down a few thoughts
would love to hear what you think
wattenberger.com/thoughts/ai…
9
10
169
19,336
this is exactly why we have horizontal tabs for workspaces in intentapp.dev
We've tried alll layouts and this one has felt the nicest in practice. We usually have a small set of things we're working on, and can tuck the rest away until we're ready.
Why horizontal tabs?
Vertical is navigation. Organizational. Hierarchical. A tree. It's "history". You don't need it on screen all the time.
Horizontal is the working set. The threads that are live. Always visible. Give it icons, colors, status. Context at a glance.
In apps like ChatGPT, Claude Code or T3 Code, horizontal space is expensive. Tabs at the top let you tuck away the sidebar and leave more room for the right panel: preview, browser, code view.
And you still see what's running, what's done, what needs your input.
5
7
275
46,793
