Bedrock Robotics

The future of construction is being built in Texas. @ChampSitePrep is betting on autonomous equipment to help their team meet demand for AI infrastructure and advanced manufacturing hubs. “The speed and scale of what's coming into this region is unlike anything we've seen before, and these projects don't wait… What Bedrock is building will multiply what our crews are capable of.” Here’s why more contractors are working with Bedrock to bring autonomy to their job sites:

Meet Eugene. Eugene is one of our hardest working machines. Eugene is also Mr Slate’s son in the Flintstones where he operates the local quarry. Coincidence? Definitely not. Autonomous every day? Definitely yes.

Full conversation: youtube.com/watch?v=3mvV9X…

Construction workers build the infrastructure that everything else depends on. They deserve careers that don't cost them their health. Omar Percy from @Sundt shares with Linda Xu why he believes autonomous technology could mean less fatigue, less pressure, and better work-life balance for the crews doing this work.

From a children's birthday party in a highway median, to a semi-truck struck by lightning, one of our team's biggest learnings from putting driverless cars on freeways is that you never know what's going to surprise you next. Robotics in the real world is like that. The environment doesn't cooperate, and the surprises don't stop coming. The job is to encounter them safely, address them, and build a system that handles whatever comes next. Construction is no different. And that's exactly why we put machines on the ground early, because the only way to find the surprises you haven't imagined yet is to show up and meet them. Field time is how we build the library of edge cases that makes the system smarter over time. If that kind of problem sounds like the one you want to work on, we're hiring.

The Bedrock Operator is how autonomy gets onto a machine. Eight cameras, LiDAR, GPS, IMUs, and a compute stack mounted to the cab turn existing heavy equipment into an autonomous system without rebuilding the machine from the ground up. Retrofit autonomy means the intelligence has to work with the machine as it is: variable hydraulics, real-world wear, the physical unpredictability of a live job site. The sensor suite we've built handles that environment, and the data it generates is the foundation for everything the system learns to do. The hardware is already out there, working, and the intelligence running on top of it will only grow from here.

Scaling compute is the easy part. The hard part is 60+ engineers submitting jobs to the same pool, all with different priorities, all expecting visibility into why their job is waiting. That's what the Anyscale Workload Scheduler solves for @BedrockRobotics. Case study: na2.hubs.ly/H05q7kb0

Before machines can work autonomously, they first need to understand the world around them. This is what a Bedrock Robotics operator sees on-site a live camera feed for human vision, and a LiDAR point cloud for machine perception. Every surface, angle, and movement is mapped in real time with millimeter-level precision the excavator’s position, bucket orientation, terrain geometry, and cut depth. This isn’t just visualization. Media : @BedrockRobotics ⚠️ This content is shared for informational purposes only. CTO Robotics Media is a media platform and does not own or develop the technology shown. Credit belongs to the original creators.

Full conversation: youtube.com/watch?v=3mvV9X…

"Why would we not want some sort of a digital operator in construction equipment?" Omar Percy, Group HS&E Manager at @Sundt, wasn't always a believer. But watching autonomous vehicles operate safely on public roads shifted something. The logic started to hold: if a digital operator can deliver consistent, repeatable outcomes in one high-stakes environment, it can probably do it in another.

@BedrockRobotics: 20K → 1.7M vCPU/hour in 12 months. 60+ job submitters. 12-person platform team. 85x compute growth. 40% cost reduction. 250+ engineering hours reclaimed. Case study: na2.hubs.ly/H05gLDd0

This is what the Bedrock Operator sees on a job site. The camera feed shows what a human operator would see. The LiDAR point cloud shows what the machine knows, the full geometry of the environment, rendered in real time, down to millimeter precision. The excavator's position, the bucket's angle, the shape of the cut: all of it mapped and updated continuously. This is the sensing foundation that makes autonomous operation possible. Before a machine can act safely and accurately, it has to understand its environment completely.

Happy Friday! Another example of hard challenges in construction. We recently did some work at a site in Texas. We had heard that the ground was hard, but had no idea that it would eat through the teeth on the bucket. Here are a few photos of the worn teeth - you can see that one of the teeth is an inch or so shorter than the others. Over a few days, these teeth wore down to nubs. They're replaceable, but of course you don't want to replace them every week - and digging gets slower and slower as they wear. We swapped tooth style, adjusted digging approach, and were back in business. From a technical pov, we have to solve this challenge in our system. The policy currently adjusts digging style to handle different conditions (assuming the right teeth). We'll build detectors for wear, and in time maybe a reasoning model to understand the environment, and what's happening and recommend changes to the machine. Every site teaches you something that simulation can't.

Before an autonomous machine can act, it has to understand. A construction site is one of the most unpredictable environments imaginable. The terrain changes with every pass. Material piles shift. Other machines move nearby. No two digs are the same. For a human operator, reading that environment is second nature, built from years of experience and constant sensory feedback. Teaching a machine to do the same is one of the hardest problems in autonomy. Scene understanding is how we're working to solve it. Our system fuses LiDAR point clouds, camera feeds, and real-time terrain data into a continuous, three-dimensional picture of everything happening around the machine. It's a living model of the environment, updated moment to moment so the machine can make decisions grounded in what's actually in front of it.

The contractors we work with aren't short on ambition. They're short on people. Our CEO, @bsofman, will be at @ENR_FutureTech next week to talk with Rachael Ferrera of @ZachryCorp and Ryan Gibson of @EclipseVentures about what autonomous heavy equipment could mean for that problem, and what it will actually take to get there. Join us to learn more about Zachry’s co-development partnership with Bedrock, and why contractor involvement now is necessary to shape the future of coordinated autonomous fleets on job sites.

Connected jobsites don’t work without people—they work better because of them. Learn how cloud tech, IoT and AI are helping contractors improve awareness and decision-making while keeping humans in control. bit.ly/4tzlFka ft. @awscloud, @DEVELONna, @BedrockRobotics

Reindustrialization is happening, just not where you think. Meet Proto-Town.

@ashleevance visited Proto-Town and got a glimpse of Bedrock’s excavators in action. Check out the video.

Building robots in the wild is fun and crazy. Lots of things happen that have nothing to do with software or ML. The day before a recent demo, a hydraulic line for the bucket on one of our machines. In that state, you can do serious damage to the machine, and of course it stalls operations. No backup except our incredible ops team who dropped everything and got in action turning wrenches. We sourced a new line, got it fitted, and had the machine running again before morning. Nothing like a midnight wrench session to remind you this isn't a simulation.