Doug Tilley retweeted
Excited to officially launch @orca_aerospace today! We are building full self-driving for satellites! Orca puts an AI operator onboard each spacecraft, so satellites can make decisions and run their missions with minimal human intervention.
Doug Tilley retweeted
Today, I am launching @ninesrobotics with @tianx2li.
We build autonomous robots for AI data centers.
Currently, AI only optimizes itself in software. By building the automation layer for server rooms, we are creating the world’s first fully software-defined, physically adaptive AI factory - the essential bridge that allows agentic AI to autonomously manage, manipulate, and optimize its own physical hardware stack.
Doug Tilley retweeted
robotics at openai is alive and well. you'd be doing yourself a favor to block any accounts that report otherwise.
15 open roles – come work with us!
openai.com/careers/search/?q…
Didn't expect to have @asim_group in an article right below the ABB/SoftBank block news.
5
203
Doug Tilley retweeted
Thanks to the teams behind LIBERO, CALVIN, SimplerEnv, RoboCasa, and RoboTwin 2.0.
Paper: arxiv.org/abs/2606.04233
Code: github.com/ripl/Manipulation…
Website: ripl.github.io/manipulation_…
Oral O3, Mon 11/9 11:30 AM · Poster 116, 5–6:30 PM
Doug Tilley retweeted
1/5 FlexiTac has been accepted to the CoRL 2026 Demo Track, and today we’re releasing FlexiTac-Gen! 🧵
Try it: flexitac.github.io/flexitac-…
Over the past two years, we’ve received many requests to customize FlexiTac for different robots. With help from AI agents, we built a tool that makes it easy to design your own sensor and export the files for fabrication.
A simple piece of tactile hardware to help move your research forward. See you at CoRL!
Doug Tilley retweeted
We’re launching Asim (YC F26) to help robotics teams find failures before deployment.
A robot completes a task perfectly. Can it handle an unfamiliar object, reason about how parts fit together, or recover when an interaction goes wrong?
We’re building standardized benchmarks for physical intelligence. We test real robots through repeatable scenarios that probe the capabilities behind successful actions: spatial reasoning, understanding physical interactions, dexterous manipulation and adaptation.
Teams can identify where performance breaks down, get actionable engineering feedback and measure whether the next software update actually improves it.
Building robots? Let’s put them to the test → asim-group.com
Asim. Benchmarking physical intelligence.
Doug Tilley retweeted
People far smarter than me are on the job of building robots.
He runs @SkildAI.
Is getting more out of AI to run robots than most others.
So I pay attention when he points out breakthroughs.
Excited to share one of the most ambitious projects from my CMU lab, led by @_tonytao_ and @JasonJZLiu.
This work shows that SOTA robot capabilities don't necessarily require massive resources. By carefully co-designing the entire robotics stack, one can achieve state-of-the-art results at low cost:
(1) Pixel: A humanoid robot under $15K.
(2) Pixel-UMI: A training framework enabling environment-aware, whole-body manipulation for dexterous and mobile tasks.
TL;DR: For under $15K, anyone can build this robot and train it with our framework to reproduce mobile dexterity results at the level of most robotics frontier labs.
We hope it is a meaningful step toward democratizing general-purpose robots capable of operating in complex, real-world environments.
Doug Tilley retweeted
Robots shouldn't need to be fine-tuned for every new task.
Introducing Runtime (YC F26). We're reimagining robot policies: explore in simulation, then act in the real world.
Watch it thread a rope.
No fine-tuning. No teleoperation.
This is a great initiative!
Always great to see more work going into evals and testing!
Introducing RoboColosseum
An open robot eval initiative by @huggingface @BitRobotNetwork @eaic_sg @DJiafei
Access:
- 100+ robots (@UnitreeRobotics G1s, @LeRobotHF SO-101s, YAMs, Franka and more)
- 140+ in-the-wild scenes
- Open rollout data
- 10,000 FREE eval hrs
Thread ↓
Doug Tilley retweeted
Introducing RoboColosseum
An open robot eval initiative by @huggingface @BitRobotNetwork @eaic_sg @DJiafei
Access:
- 100+ robots (@UnitreeRobotics G1s, @LeRobotHF SO-101s, YAMs, Franka and more)
- 140+ in-the-wild scenes
- Open rollout data
- 10,000 FREE eval hrs
Thread ↓
Chasing better physics solvers and new simulation frameworks will get us 90% of the way there, but the only real test of a robot's capabilities, and safety, is in the physical world.
As this survey by @EliJalbout et al. points out, policies love to exploit simulator corner cases that become dangerous the second they touch real hardware: arxiv.org/abs/2510.20808
Robotics could become an $8T market by 2035.
One serious accident could shake public trust and slow adoption across the industry.
The International Robot Safety Conference is on the next 3 days, Nov 3–5. Getting safety right is how we make that future possible.
2
5
334
"The scores looked fine. The traces didn't."
Had a great time speaking at the Bench Summit! With agentic robotics being such a hot topic right now (distinct from WAMs, VLAs, or pure RL), it’s awesome to see how much energy the community is putting into evaluating these systems and opening them up to the broader public.
Top-line scores are a great start, but the most exciting insights are hiding in the execution traces. Digging into the traces together is how we turn benchmark success into real-world reliability. Always read your traces!
The picture shown is a recreation of one of the slides presented at the opening.
1
5
271
Doug Tilley retweeted
Replying to @_varunnair
@_varunnair and I spent the last few weeks mapping the robotics industry.
230+ companies across the stack: data, simulation, models, robots, hardware, and deployment.
Missing someone or got something wrong?
Check it out and let us know at:
roboticsmap.org/
This Wednesday I’m speaking at Bench Summit (@kaggle , @GoogleDeepMind & @zapier) about physical robot evals.
What can we learn from LLM/VLM benchmarks, and what do success rates miss?
I’ll share our perspective at @asim_group . Come say hi!
Doug Tilley retweeted
Today we are launching @ObsessionDB, the realtime data warehouse for agents.
Your agents are hungry for data. Feeding them is getting expensive.
Your data warehouse wasn't built (or priced) for that.
We learned this the hard way. We scaled our last company to petabytes on ClickHouse and watched the infra bill go vertical.
So we built @ObsessionDB . The ClickHouse engine, rebuilt with compute and storage separated, so your infra scales with your agents instead of your bill.
100x faster and 20x cheaper than Snowflake and Databricks. Already live, processing trillions of rows a day.
Agents are just getting started...


















