Twitxy
Saved
@yogesh_origin's profile picture
View on X

Yogesh Ghaturle

Verified account This account is verified. Learn more

@yogesh_origin

CEO & Founder @origin_autonomy | Building Time Machines for Construction!

Ex-Founder - Perpule (Acq by Amazon) | AI x Robotics | ACM-ICPC World Finalist
San Francisco, CA origin.tech Joined March 2015
2.2K Following 760 Followers 569 Posts

Latest posts 5

Yogesh Ghaturle Verified account @yogesh_origin
Thrilled to announce that @origin_autonomy has raised $20M in total funding, co-led by @Accel and @ZacuaVentures.

Excited to be building this with @ashishxdaga, @tusharmakkar08 and the super talented team at @origin_autonomy. We’re building general-purpose robots for interior construction.

Special thanks to @johnkoetsier at @Forbes for covering our journey.

News Article: tinyurl.com/ym2vj7ts
45 replies 40 reposts 247 likes 29K views View on X
Yogesh Ghaturle Verified account @yogesh_origin
this is the hidden stack behind every serious deployment.

but I would organize it around incidents, not tools. a field failure should become one reproducible artifact: synchronized robot + environment state, software/model/hardware versions, operator intervention(if any), recovery path and deterministic replay.

collecting logs is table stakes. the compounding advantage is turning one customer failure into a regression test and then a permanent improvement across the fleet.

we have built something similar at @origin_autonomy
Joe Harris Verified account @_joe_harris_ Aug 30
things every serious robotics team ends up rebuilding once the robots leave the lab:

- the sim → test → logs → replay → eval loop for policies
- observability across ros 2, CAN, fieldbuses, video and sensors that ops can actually use
- fleet management thats slightly more duct-taped than pure ssh + spreadsheets
- a data layer tying missions, bags, firmware versions, incidents, labels and models together
- audit-grade safety cases, traceability and operating procedures
- tools that let support and field teams diagnose, teleop and recover without opening a terminal

robotics won't have its akeoff moment until each of these have off-the-shelf solutions that just work.
View quoted post on X
0 replies 0 reposts 17 likes 551 views View on X
Yogesh Ghaturle Verified account @yogesh_origin
tasks are model evaluations. workflows are customer evaluations.

customers don’t pay for tasks. they pay for completed workflows.

folding a napkin is a task. keeping a dining room supplied with 1,500 table-ready napkins throughout a shift correctly folded, stacked and ready for staff is the workflow.

a robot can succeed at every individual task and still fail the operation because the handoffs between those tasks don’t work.

robotics needs to move beyond task success rates and measure workflow completion: usable throughput, quality, recovery and human intervention.

Congratulations to team at @DynaRobotics for crossing that chasm!
Dyna Robotics @DynaRobotics Aug 27
After so many demos, models, pilots, robots are still struggling to land real deployments with real customers. Until now.

Today we’re excited to share that our robots have successfully crossed the ROI threshold, and Din Tai Fung, one of the highest revenue per location restaurant chain in the US, is rolling out Dyna robots across its extensive restaurant network.

This brings our rollouts across hotels, logistics, data centers, and many other use cases to a fleet that reaches hundreds of robots by the first half of 2027. And we’re just getting started.

It’s been a wild year, and today we’re double clicking on the battlefield stories and sharing a few learnings about scaling robot deployments. We are just scratching the surface.

Read the full blog post: dyna.co/news/scaling-c…
Video preview
View quoted post on X
0 replies 0 reposts 15 likes 394 views View on X