fig (@figbrains) header
fig (@figbrains) profile picture

fig

@figbrains

hyperintelligence for humanity

 San Francisco 🌁
fig.inc/
Joined January 2024
  • Tweets111
  • Following5
  • Followers215
15Media
Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)Photo by fig (@figbrains)
Pinned Tweet
fig (@figbrains) profile picture
fig
@figbrains22h
We're releasing RealCUA, the first computer-use dataset of open-ended work in evolving environments. 600+ hours (and growing!) of real multi-objective work with interleaved tasks across thousands of steps in realistic desktops. 11-hour sample & access: data.fig.inc/realcua
20
37
8
401
32,309
Replying to @HarshSikka
@HarshSikka Harsh, this is awesome work! Generating high-quality, high-fidelity data in this way helps both, build better harnesses (to cover what is not in the models) and to build better models over time. Well done! @figbrains
1
1
4
134
fig retweeted
Cosign (@cosign) profile picture
Cosign
@cosign14h
What products launched today? - @OpenAI introduces GPT-6 Cyber, a cybersecurity research model available to select Daybreak Red customers. - @MistralAI previews managed agents and apps in Mistral Studio. - @harvey previews fact graph, a tool for disputes, litigation, and investigations. - @perplexity_ai launches Teammates, proactive personal AI agents. - @reductoai launches r-1 mini, a document parser priced at $1 per 1,000 pages. - @raindrop_ai launches the public beta of Simulations, a tool for testing AI agents against production scenarios. - @arcjet launches Himalia 1, a model for detecting prompt injections. - @Tiny_Fish launches TinyResearch and TinySet, tools for web research and dataset creation. - @PatronusAI open-sources GlobeBench, a benchmark for LLM simulation of reinforcement-learning environments. - @supermemory launches preprint, a tool exposing browser tabs as Markdown files. - @distcog unveils VINCO, a reinforcement-learning model for 1v1 Catan. - @figbrains introduces RealCUA, a dataset of 600+ hours of desktop work across 35 apps, available on request. Follow tech industry news on Cosign: cosign.co/feed
Image from Cosign (@cosign)
10
8
3
40
2,492
2/ Full workspace video Recordings cover the entire desktop with multi-monitor support, on persistent desktops. Every task starts in a workspace shaped by earlier work, and information moves across app and window boundaries.
Image from fig (@figbrains)
1
6
337
3/ Open-ended trajectories Operators carry out their own work on personal machines, without assigned tasks or scripted instructions. Multi-thousand-step workflows, with the longest continuous trajectory running over 12 hours. 87% of hours come from recordings an hour or longer.
4
273
Our latest writeup shares what hundreds of hours of real, open-ended work taught us about the learning signals missing from today’s computer-use data. We'll be sharing new datasets and environments in the coming weeks, reach out for an early preview!
Understanding Open-ended Work from a Data Perspective

Understanding Open-ended Work from a Data Perspective

Article · @figbrains
2
4
6
942
Join the Fig team tomorrow at Asilomar SSC! Keep up with our research here: info.fig.inc/r/7RGXaZ nitter.gq/Yangyue_Wang/status/21…
locke (@Yangyue_Wang) profile picturelocke
@Yangyue_WangOct 26
Headed to the Asilomar Conference on Signals, Systems and Computers tomorrow! I'll be there for @figbrains and @ManifoldRG, presenting GUI-Perturbed. If you're into agents, continual learning, memory, self-improvement, or multi-agent learning, come say hi. I'd love to grab a coffee and talk! Poster TP8a2.24, Tue 1:30pm, Merrill Hall Technical report: fig.inc/blog/domain-randomiz… Paper: arxiv.org/abs/2604.14262 #Asilomar2026
Domain Randomization for Computer Control

Domain Randomization for Computer Control

GUI models that score 90%+ on benchmarks break under trivial UI changes. We built GUI-Perturbed to measure robustness, not just peak accuracy.

fig.inc
4
206
I'm really excited about the upcoming data technologies the Fig team is developing. We're systematically solving the open data problem. 𝗜𝗳 𝘆𝗼𝘂'𝗿𝗲 𝘄𝗼𝗿𝗸𝗶𝗻𝗴 𝗼𝗻 𝗮𝗽𝗽𝗹𝘆𝗶𝗻𝗴 𝗮𝗶 𝘁𝗼 𝗹𝗼𝗻𝗴 𝗵𝗼𝗿𝗶𝘇𝗼𝗻 𝗼𝗿 𝗼𝗽𝗲𝗻 𝗲𝗻𝗱𝗲𝗱 𝘁𝗮𝘀𝗸𝘀 𝗶𝗻 𝗿𝗲𝗮𝗹𝗶𝘀𝘁𝗶𝗰 𝘀𝗲𝘁𝘁𝗶𝗻𝗴𝘀, 𝗹𝗲𝘁'𝘀 𝗰𝗵𝗮𝘁! Many frontier teams (including us) have run into this big hairy problem of extracting useful signal and building effective, expressive environments for the kind of work someone might do over the course of a full week in a high performance job. We're building a science to solve this. data.fig.inc/preview?utm_sou… nitter.gq/figbrains/status/21136…
Fig · Solving the Open-Ended Data Problem

Fig · Solving the Open-Ended Data Problem

New scalable approaches that turn real open-ended, multi objective knowledge work into signals to train frontier AI systems. Request an early preview.

data.fig.inc
fig (@figbrains) profile picturefig
@figbrainsOct 23
Real world work is open-ended: multi-objective, interleaved, and noisy. Extracting good learning signals from this has been intractable. We're developing new scalable approaches that turn real open-ended, multi objective knowledge work into signals to train frontier AI systems.
2
3
8
61,184
Real world work is open-ended: multi-objective, interleaved, and noisy. Extracting good learning signals from this has been intractable. We're developing new scalable approaches that turn real open-ended, multi objective knowledge work into signals to train frontier AI systems.
2
1
4
10
62,818
We're sharing early previews with teams ahead of our upcoming releases. If you're deploying models into complex, open-ended environments, we want to work with you: data.fig.inc/preview?utm_sou…
Fig · Solving the Open-Ended Data Problem

Fig · Solving the Open-Ended Data Problem

New scalable approaches that turn real open-ended, multi objective knowledge work into signals to train frontier AI systems. Request an early preview.

data.fig.inc
2
95
The Fig team will be in Boston for the ARC Prize summit! Reach out to get a sneak peek at the next generation of intelligence we’ve been quietly building!
harsh✌️ (@HarshSikka) profile pictureharsh✌️
@HarshSikkaOct 22
I’m attending & presenting @arcprize summit in Boston. Will be around tomorrow & Saturday. Dm / reply to me the most cracked people you know, i want to brain drain everyone (researchers, engineers, founders investors anyone) I’ll let them brain drain me too, so it’ll be worth it
2
2
104
Benchmarks hide serious gaps in model performance. Last Friday, we learned the extent of this in "How Good Are Frontier Models, Really?" @Yangyue_Wang from @figbrains dug into why model capabilities remain so jagged even as benchmark scores keep climbing. Missed it? No problem! Watch the full talk here: youtube.com/watch?v=59WZlXqH…
Frontiers: How good are frontier models, really?

Frontiers: How good are frontier models, really?

Welcome to Frontiers, a series where we bring researchers, engineer...

youtube.com
1
7
1
9
297
A study from the team showing Astra's and Opus 5.5's jagged reasoning behaviors in simple puzzles. More data points on frontier models' failure profiles. nitter.gq/figbrains/status/21100…
fig (@figbrains) profile picturefig
@figbrainsOct 13
A few weeks ago, we reported that frontier models make less progress as simple 2D mazes get longer. We tested GPT-6 Astra and Claude Opus 5.5 - length doesn't matter, but they struggle with reasoning about simple obstacles without being told.
1
2
148
fig retweeted
dip (@dippogriff) profile picture
dip
@dippogriffOct 13
Cranking up reasoning doesn't solve all problems. More thinking more may generate better text or artifacts, not necessarily actions.
fig (@figbrains) profile picturefig
@figbrainsOct 13
A few weeks ago, we reported that frontier models make less progress as simple 2D mazes get longer. We tested GPT-6 Astra and Claude Opus 5.5 - length doesn't matter, but they struggle with reasoning about simple obstacles without being told.
1
4
130
A few weeks ago, we reported that frontier models make less progress as simple 2D mazes get longer. We tested GPT-6 Astra and Claude Opus 5.5 - length doesn't matter, but they struggle with reasoning about simple obstacles without being told.
1
4
4
9
11,402
more replies
That is the gap we care most about at Fig: systems that improve from the consequences of their own actions. Next we put the same mazes in a 3D world and in plain language, with full history, to separate familiarity with the interface from understanding of the task.
1
2
21
Built with @ManifoldRG by Pranav Guruprasad, Sean Rivera, Helen Lu, Arushi Jain and Harsh Sikka.
2
23
Load more