CEO for @quint_lang because I care about software too much | Born and living in Brazil 🇧🇷
Joinville, Brasil
- Tweets 1,804
- Following 1,187
- Followers 1,259
- Likes 25,978
Pinned Tweet
We considered many shapes for the Quint product. We experimented a lot. It's time to put something out into the world!
Waitlist link below
5
4
53
4,942
Help us build this and sorry in advance if we got something wrong!
We put together a practical reference to formal verification tools:
quint.sh/guides/formal-verif…
If you're getting started, use it to explore the landscape.
If you're already an expert, help us make it better.
If you find a missing tool or any mistake, please tell us! We want this to be an accurate, useful, and current reference.
1
10
522
For those of you who like TLA+, we can call it TLA+ Studio if you prefer!
(Quint transpiles to TLA+. Video below is a mock, we don't really have TLA+ transpilation in the product)
3
7
70
4,319
This is obviously AI generated but it's actually a really really good explanation that matches my mental model for TLA+/Quint in many ways.
(The Lean part was a bit weird, I won't try to comment on it tho).
Quint Studio is doing the stuff explained here, reproducing counterexamples and helping you fix it but also adding test scenarios for correct but untested paths found from the model.
The auto-formalization inside Studio has amazing techniques not explained here tho, so stay tuned for that.
> This seems compelling, but I don't know much about formal verification. Read Boris's tweet and create a video explaining the concept.
2
35
4,322
Gabriela Moreira retweeted
Babe are you okay? You've been saying formal methods is the future all week but didn't even tried the 5-minute Quint getting started tutorial.
1
4
23
769
Gabriela Moreira retweeted
Seeing a whole lotta tweets about how the era of handcoding is over and no tweets about people ricing their agent harnesses
C'mon, if people customize VSCode they should be customizing Claude Code. Show me your hooks, your conversation DAG visualizers, your weird slash commands
38
4
119
8,797
Gabriela Moreira retweeted
Quint Studio has been in beta a couple weeks now. Here's a quick breakdown of what it does and where we're at:
1
3
5
275
Want to get better at (or started with) formal methods? Surround yourself with experts @hillelogram @DominikTornow @k0nn0v @muratdemirbas @lemmster @ilyasergey @ankushpd @Leonard41111588 @MarcJBrooker @vanlightly @immadnaseer
You're welcome.
3
7
44
5,566
. @heyyfernanda I wanted to tag you but couldn't find you here! My bad!!
1
1
219
I used to do this back in 2019 when I had to ssh into staging and prod all the time! Staging was green and prod was red. Big piece-of-mind value.
(I've been building and chilling piece-of-mind tools ever since)
my terminal automatically tints when i ssh into something
one of 32 unique tints is deterministically chosen based on the remote's ssh fingerprint
no more silly accidents
13
750
Gabriela Moreira retweeted
Incredibly proud of the Cycles Pay team for this, was a huge lift.
We're the first & only privacy app on Arc right now, until they launch their privacy sectors.
Try it out pay.cycles.money/
Send invoices, pay bills, and put your money to work, all privately onchain
Cycles Pay is now officially running on @arc mainnet.
This is the first application built on top of the Cycles protocol, a privacy-preserving clearing algorithm that finds loops of mutual debt and settles them without altering what each party is owed on net.
Try it today at pay.cycles.money/
We already have a web app and a mobile app, and we’ve already added support for both businesses and individuals.
Perhaps more importantly, it has a combination of features no one else in the industry has built yet: private stablecoin payments, earn vaults, invoicing, expense splitting and AI assistants.
And we’re working on adding invoice financing, credit scores, balance lending, and rotating savings groups with the ultimate goal of becoming the first on-chain credit bureau to deliver credit scores based on real payment history and active payment obligations.
@arc is the foundation. We chose it because it is designed for payments with USDC as gas, sub-second finality, and reliable global infrastructure.
10
11
84
7,581
Gabriela Moreira retweeted
what it's like to wake up and realize the whole world is talking about formal verification
5
28
841
Gabriela Moreira retweeted
Formal specifications are great for bug finding, but that's not all.
Quint Studio uses the same specs that checked for bugs to assess quality of test scenarios and more.
Join the beta: bit.ly/4wr3FK9
I used Opus 5.5 to formally verify the Claude Agent SDK using Lean. A couple short prompts = 16 PRs fixing various bugs and race conditions. Video attached.
TLA+ also works well. I sometimes combine Lean and TLA+ to look for issues around data flow, concurrency, and state mgmt.
I don't know either language well, but Claude is excellent at both. This approach is super useful for formally modeling your code and finding bugs that a human probably wouldn't have spotted.
Is formal verification the future of coding (or at least, bug finding)?
4
3
11
744
Gabriela Moreira retweeted
There are a lot of people dunking on this. I have been critical of many things the Claude Code team has said in the last 6 months but this is not one of them.
Folks should use the excitement to teach and guide those incoming. Discuss pitfalls and why you can't formally verify all software. Show what is possible, explain what isn't.
I understand it can be overwhelming but that's the best outcome for everyone. This trend is much more valuable to us as an industry than "markdown files will describe how my system behave".
I used Opus 5.5 to formally verify the Claude Agent SDK using Lean. A couple short prompts = 16 PRs fixing various bugs and race conditions. Video attached.
TLA+ also works well. I sometimes combine Lean and TLA+ to look for issues around data flow, concurrency, and state mgmt.
I don't know either language well, but Claude is excellent at both. This approach is super useful for formally modeling your code and finding bugs that a human probably wouldn't have spotted.
Is formal verification the future of coding (or at least, bug finding)?
20
18
239
42,038
Gabriela Moreira retweeted
chat; watch this talk. piped.video/watch?v=9O6T-iyC…
I used Opus 5.5 to formally verify the Claude Agent SDK using Lean. A couple short prompts = 16 PRs fixing various bugs and race conditions. Video attached.
TLA+ also works well. I sometimes combine Lean and TLA+ to look for issues around data flow, concurrency, and state mgmt.
I don't know either language well, but Claude is excellent at both. This approach is super useful for formally modeling your code and finding bugs that a human probably wouldn't have spotted.
Is formal verification the future of coding (or at least, bug finding)?
2
5
48
6,300
Gabriela Moreira retweeted
ちょっと寄り道して Alloy 6 で入ったという temporal logic の機能を見ている。ところどころ違う部分があって面白い github.com/hibariya/tla-exam…
2
3
226
Running a very similar experiment today with Quint Studio. My demo is not so impressive yet, but the techniques are solid.
Also, y'all can read and play with the Quint specs if you want to understand what they are doing. Maybe that's not really the case for TLA+ and Lean.
I used Opus 5.5 to formally verify the Claude Agent SDK using Lean. A couple short prompts = 16 PRs fixing various bugs and race conditions. Video attached.
TLA+ also works well. I sometimes combine Lean and TLA+ to look for issues around data flow, concurrency, and state mgmt.
I don't know either language well, but Claude is excellent at both. This approach is super useful for formally modeling your code and finding bugs that a human probably wouldn't have spotted.
Is formal verification the future of coding (or at least, bug finding)?
1
4
22
1,225
Gabriela Moreira retweeted
New video on how to trust AI-generated code before you ship it. Your agents and software factories need tools that enforce your intentions.
See what that takes: bit.ly/4y91lsf
1
4
4
443
Never delegate understanding²
(From my BugBash talk)
Delegate coding. Never delegate understanding.
If code was previously your source of truth for managing understanding, then you need a new source of truth artifact. If code-centric workflows were how you developed that understanding, then you need new workflows.
1
1
11
423