The agent harness to accelerate physical AI.
Joined November 2022
- Tweets71
- Following24
- Followers23
- Likes31
"We have multiple generations of robots and multiple generations of autonomy software in the field at the same time, and each one evolves on its own timeline."
That's Jackie Song, Senior Systems Engineer at Serve Robotics, on what engineering looks like at fleet scale.
We built Trinity to help engineers tackle humanity’s greatest challenges and bring Physical AI to scale. Trinity is an AI agent harness, with the engineer at the center of the story.
That's why this video was made on physical tape without any AI involved.
Mark your calendars! On October 8, we're showing Trinity live.
Trinity holds the engineering data together: requirements, tests, design parameters and variants recorded and related in one graph, with rules and agents reasoning over it.
Register here: trace.space/events/webinars/…
Janis Vavere @thehotguyintech for @tech_eu on the next bottleneck in hardware engineering: as AI makes it faster to build and test hardware, the hard part becomes keeping track of what is being built, and how to manufacture it at scale.
Read more: tech.eu/2026/10/01/tracespac…
Trinity makes scaling physical AI possible.
Record: all engineering data in a single graph.
Relate: connect every requirement, test, parameter and variant.
Reason: change impact analysis across the whole graph.
Learn more at: trace.space/trinity
Trace.Space launches Trinity: the AI harness that makes scaling physical AI possible.
Hardware teams have the ideas, the funding and working prototypes. What slows them down is finding out, weeks too late, what a change broke.
See Trinity live on October 8.
Our founders, Janis Vavere (@thehotguyintech and Karlis Broders, have worked on this problem for more than a decade. They built Trace.Space on one connected graph from day one, knowing AI would eventually be able to do real engineering work with the right foundation.
Tomorrow, we're launching our answer to the last unsolved problem in hardware engineering.
What we've built at Trace.Space makes systems engineering work the way it was always supposed to.
Stay tuned and make sure you don't miss it.
The future you were promised – the humanoid robots, the rockets that land themselves, the autonomous cars – all of it already exists. It exists as prototypes, as demo units on a closed track, and production runs you could count by hand.
They get the contract, and then they hit the wall. Inventing the machines was the achievable part, but making the hundred thousandth unit as good as the first is where programs stall, because the software changes weekly and no two units in the field stay configured the same way
Starting this week, you can choose how you want to work in Document view: keep it compact, or show attribute titles at a glance.
We’ve been adding a lot more flexibility to how Trace.Space looks and feels. This is just one of those improvements.
Trace.Space retweeted
The future does not exist. Yet. We have all been promised the embodied AI - the autonomous robots, the rockets that land themselves, the autonomous cars. They exist as fancy demos, as prototypes, as small production runs. But what about scale?
Scale is impossible without @tracedotspace. This is the team that is enabling hardware companies to make the future a reality.
What we’re working on right now is built for where hardware is going next: systems engineering for products that never stop evolving. Stay tuned.
Most hardware companies in the world are building software-defined machines that evolve with every mile, every delivery, every deployment.
More on this in the thread ->
This creates an opportunity to rethink how we define and manage the system itself. Instead of requirements describing one product at one point in time, they can evolve alongside every configuration, parameter, test, and software update.