It is not clear to me how or if /teamwork is any different than Claude dynamic workflows.
Antigravity as a harness feels very behind. There's not even any auto mode. And they announced /goal long after I ever felt the need to use it in CC or Codex.
As noted by others, we already announced parts of /teamwork in Antigravity here: https://antigravity.google/blog/teamwork-when-ai-becomes-a-r... with more visualizations and more. Thi post was meant to report some advances in math/science/coding/engineering to show-case examples of tasks you can do with it.
We are constantly improving/evolving /teamwork and pushing those changes for external users. Will also post more about improvements and details for building teams of agents dynamically as Antigravity blog posts soon. Stay tuned!
Here is a short list of example results reported on the post; again we will have dedicated Antigravity posts for some of these.
* Math Open Problems & Formal Verification: Reporting 10+ open math and CS theory problems solved including 50M-line Lean 4 proof of Courtade–Kumar conjecture plus progress on several open problems in quantum query complexity, Tarjan's Deque Conjecture.
Also autonomous Lean formalizations of the Weil (1.8M lines), Birkhoff, and 3D Bochner–Riesz conjectures via Teamwork-LEAP, plus updates on Deep Review for STOC’27.
* Chemistry & Discovery: 95% top-1 exact match on NMR molecular structure elucidation (nearly 2× single-agent). Also empowering Nuclear Fusion and Quantum algorithm results as well autonomously generating a CausalAnalyst Benchmark. In addition, we show how /teamwork can start from Alphazero paper, implement a competitive go game, and autonomously improving it by playing with KataGo to get to the medium Elo level.
* Several Systems, TPU Kernel, & Infra:
* Clustering 1-Trillion-edge graphs on FarArray and RDMA (up to 180× faster),
* 800K-line Zircon & 170K-line libgav1 safe Rust rewrites (20–30% faster than C++/ASM),
* 4.09× JAXBench TPU kernel speedups with DuoDiscover,
* Google internal examples: surpassing 1 PiB fleet memory savings, and 24% fewer Borg TPU preemptions for Borg Scheduling.
I think the authors also used Antigravity to write this article, because it sucks at communicating whatever point it’s trying to make. I think it’s something like “We did the OpenAI 100 problems thing, but across many disciplines, and in collaboration with the experts instead of just dumping results on the world, and we also made the tooling feature available,” but I’m not sure because it has that distinct Gemini flavor of packing in so many low-level details that the actual point becomes entirely obscured.
It is a shame that Gemini prose has become too detail dense in recent releases. Its prose was one of its distinct advantages over Claude with that annoying Claude speak...
It seems like they already announced /teamwork earlier [0], but now just announced that they got some scientific results out of it. They probably intentionally want to keep this low key, but in that case it's not clear why announce it at all.
It sounds like they're saying they now offer an option, "/teamwork", that will throw more computer at a problem by choosing a team and a workflow that you the user don't have to (or get to?) know about.
Interesting that a Rust re-write pre optimisation was 30% slower than the C++ equivalent.
I wonder if they optimised the C++ it would actually have more speed benefits than the Rust alternative. Only they can't claim it's a Rust rewrite like every other company at the moment.
It's pretty clear "credit" for things is going away, we're going to have to adapt to that. This includes people achieving big things with or without AI, don't expect anyone to be impressed or care, you just have to feel good about what you did being used.
That has been true of a lot of good science for a while.
The person who gets the credit is just the person who wrote the grant. They are usually _capable_ of doing good science but the demands of being a PI mean they are mostly just grant writers and administrators. Huge teams of unknown postdocs, grad students etc. are doing most of the work.
That's inside that particular singularity, where nothing means anything, you just "adapt".
But from the outside, it's just a bunch of people turning into a blob we don't care about as a whole. We still care about the individual humans who haven't jumped into that gravity well. Humans care about their ancestors, their neighbours, authors and audiences. The people who don't can't fathom that, it's a sort of blind spot for them, but we are completely untouched and unmoved by that. The ideas in your head about the bonds between other people doesn't affect those bonds.
Antigravity as a harness feels very behind. There's not even any auto mode. And they announced /goal long after I ever felt the need to use it in CC or Codex.
We are constantly improving/evolving /teamwork and pushing those changes for external users. Will also post more about improvements and details for building teams of agents dynamically as Antigravity blog posts soon. Stay tuned!
Also I added a list of examples here: https://x.com/mirrokni/status/2121351383236436247?a=ma9UPZg_... and https://lnkd.in/p/esVTpR6p. I'll add a comment with examples as well.
* Math Open Problems & Formal Verification: Reporting 10+ open math and CS theory problems solved including 50M-line Lean 4 proof of Courtade–Kumar conjecture plus progress on several open problems in quantum query complexity, Tarjan's Deque Conjecture.
* Chemistry & Discovery: 95% top-1 exact match on NMR molecular structure elucidation (nearly 2× single-agent). Also empowering Nuclear Fusion and Quantum algorithm results as well autonomously generating a CausalAnalyst Benchmark. In addition, we show how /teamwork can start from Alphazero paper, implement a competitive go game, and autonomously improving it by playing with KataGo to get to the medium Elo level.* Several Systems, TPU Kernel, & Infra: * Clustering 1-Trillion-edge graphs on FarArray and RDMA (up to 180× faster), * 800K-line Zircon & 170K-line libgav1 safe Rust rewrites (20–30% faster than C++/ASM), * 4.09× JAXBench TPU kernel speedups with DuoDiscover, * Google internal examples: surpassing 1 PiB fleet memory savings, and 24% fewer Borg TPU preemptions for Borg Scheduling.
[0]: https://antigravity.google/blog/teamwork-when-ai-becomes-a-r...
The pattern seems to be pervasive at Google. Too much Ai, not enough human, top-down orders
I wonder if they optimised the C++ it would actually have more speed benefits than the Rust alternative. Only they can't claim it's a Rust rewrite like every other company at the moment.
That's OpenAI (allegedly), not google
Skype and Hangouts were forbidden for sensitive research at some German universities:
https://www.cms.hu-berlin.de/de/dl/multimedia/bereiche/tele/...
Now everyone leaks everything by using AI even for mundane things like file format conversions.
The person who gets the credit is just the person who wrote the grant. They are usually _capable_ of doing good science but the demands of being a PI mean they are mostly just grant writers and administrators. Huge teams of unknown postdocs, grad students etc. are doing most of the work.
But from the outside, it's just a bunch of people turning into a blob we don't care about as a whole. We still care about the individual humans who haven't jumped into that gravity well. Humans care about their ancestors, their neighbours, authors and audiences. The people who don't can't fathom that, it's a sort of blind spot for them, but we are completely untouched and unmoved by that. The ideas in your head about the bonds between other people doesn't affect those bonds.