Building ref.tools, previously @figma & @PalantirTech
San Francisco, CA
- Tweets 529
- Following 321
- Followers 1,644
- Likes 824
Ref September Update
You don't need to visit github․com anymore.
👀 Review diffs next to the ref.
✅ Approve a ref and forward that approval to PRs.
Also, bundle size ⬇️ and cache hits ⬆️ so everything is way faster!
4
254
this is what a ref is
it's the notebook of everything you've done to explore and solve a problem.
it's easy to share with your team
it's basically impossible for someone to just "show you their prompt" now, because everything is about references, skills and examples
I often ask my agent to look at 3 other repos I've made first, search the web for references, use other AI APIs, etc.
2
9
930
this is the fundamental ai content problem:
if you spend a lot of time looking at decks, its easy to tell which ones are thoughtless
if you rarely engage with slide decks, the ai generated ones will look amazing with zero thought
the timeline is full of it in every medium (design, slides, games, video etc)
if your audience is at all sophisticated in the medium you're producing, you can't skip thinking
140
I'm changing how i review my teammate's work (code or plan)
it used to be i'd mostly read, react and throw out questions to start discussion and prompt the owner's research. I was a good collaborator by identifying problems or nudging in better direction
now it's the opposite.
If i'm leaving comments and pulling your agents out of flow, the comments deserve more weight. Gone are the days of "did you consider <X> approach?". The barrier to answering that myself is now so low that I owe you better
the goal is still not to gatekeep or take ownership of my teammates work. the bar for a good review is simply higher now that I can research, riff and test with a prompt
2
151
the designers+machines meetup last night was so good (s/o @ben_issen)
i go to a lot of ai meetups and the vibe was noticeably different than most. it was cheerful and creative and curious
if you're in SF and like design and ai you should try to get in on the next one
2
7
328
i skipped worktrees and went straight to cloud agents last summer thanks to @terragonlabs (rip)
these days whenever i see someone using worktrees it makes me a bit sad because their life could be just a bit easier
5
287
the shift away from tickets for managing a team's work is very real
i still need visibility and accountability.
but we don't need to decompose a project into parts. we used to do that to get better estimates and so multiple people can work on it in parallel. neither is relevant anymore
what's working really well for us is individuals in lanes and they commit to "I'll deliver<X> by next week"
we do plan review to stay aligned and daily stand-up to share ideas and get feedback
For the first time, I think issue tracking is dying. At least for small teams.
After five years, I shut down our Linear workspace.
Every issue I'd file now goes one of two places.
If it's urgent enough to deserve tracking, it's urgent enough to hand to an agent right now. It comes back as a PR, not some card rotting in a backlog.
If it needs more product thinking than that, it belongs on our feedback board.
The issue tracker felt like the middle ground between them. I don't think that middle ground exists anymore.
These days my agent workspace is the tracker. What's running is what's being worked on. Nothing else needs a status.
This gets harder with a bigger team, I know. But these days, I'm seeing similar across my friends.
1
3
669
that makes a ton of sense to me!
i'm not as familiar with A2A but having a tool call backed by some form of intelligence is something we do a few different places in ref.tools
we have a hosted claude code as part of the app so there's UI for launching it and the http mcp server has a `launchAgent` tool. Now we can have that same tool in the browser
I don't imagine this as the main workflow but as more sites adopt webmcp, sites will just feel broken when you can't prompt the core functiosn imo
2
95
oh no ref totally supports all the levels you'd expect from a doc. personal, team, email and link access, viewer/commenter/editor etc
1
2
49
i just realized one really important anti-design slop force we lost
there used to be this thing where a designer would miss a use case or error state but then some poor engineer would be tasked with programing it. they'd actually have to manually, by hand, type code for each of these case.
this was always so incredibly annoying. you'd get a beautiful golden path and then a 15 unhandled edge cases.
my favorite i ever got was "this reminder shows at exactly the right moment" 😅 like okay what data we using for that? how do we define the right moment? are you trying to fight me?
this was frustrating for the designer too bc they sweat over this perfect design and the engineer is only concerned with silly edge cases (note: good designers do care about edge case. many are not good)
having to type every edge case required us to think about UX. this was annoying for everyone BUT it was positive frictionnand it was a force for good in the universe
5
11
1,605
This is an excellent break down of multiplayer-AI!
At ref.tools we thinking about multiplayer-AI for coding as multiplayer context engineering.
One important requirement is that the shared context must be observable and mutable to the entire team.
The implication is that is basically impossible if you are sharing agent sessions directly. For non-trivial work, that's like dropping into a massive 100+ message slack thread and trying to build a mental model for what's happening.
You need surface that makes it easy to explore your ideas in-depth with and agent AND easy for your teammate to glance at and provide feedback.
It needs to be human-legible and agent-useful.
6
1
42
11,099
i like working with vivek
We do almost no code review.
Most of our review happens in the planning process with ref.tools and Fable, not at the PR stage.
After we’ve aligned on key decisions/tasks as a team, I use Ref’s built-in orchestrator to fan out multiple tasks in parallel to Opus 5.
Once the initial implementation is done, the PRs all go through several AI-driven reviews to ensure:
* The implementation matches the intent of the plan (Ref conformance checks).
* General bugs are caught, fixed and noted in the plan (Codex is great here)
* Security review (Codex also great)
* Linting, unit and integration test suites are green (standard CI practices)
The orchestrator agent is typically still managing all of this when it’s in-flight. It notifies me if anything needs my input.
We also have preview deploys on all PRs so we can always manually verify features end to end ourselves when needed.
2
492
wait a sec if everything's a web app and we all work with a side car agent via webmcp
...is the browser the governance layer?
3
5
493
nice! we are too @ref_tools haha
but specifically for developers. maybe we should chat!
1
3
108