Install on TestFlightWatch the XRCC trailer
XRCC 2026 Finalist

AI copilot for field technicians

Record one expert demo. Infinite real-time coaching via smart glasses or mobile.

Install on TestFlightBook a pilot
The problem

Your senior technician walks out the door. Decades of judgment walk out with him.

Half of process knowledge is held in people's heads. Roughly 42%McKinsey is tribal, undocumented, lost the day they leave. The 600-page service manual that's supposed to replace them is exhaustive and unreadable in the moment.

Junior technicians inherit the gap anyway.

  • The Gap

    25% of failed dispatches trace directly to a training gap.

  • The Cost

    The average repeat truck roll costs $300 in fuel and time, before counting customer downtime.

  • The Crunch

    The US manufacturing workforce is 44.8 years old on average, the oldest it has ever been. By 2033, we face 3.8M jobs with roughly half left unfilled.

Record once. Train everyone. Solve everything.

01 / CAPTURE

Record an expert demo

Put on the glasses or hold up your phone, then talk through the task once. The AI automatically breaks your video down into structured, step-by-step instructions and clips each step into a short video snippet.

REC · 00:07
Extracted SOP
Install structural blind rivet nut insert
  1. 01Select mandrel and nose piece
  2. 02Set tool stroke depth
  3. 03Drill and deburr pilot hole
  4. 04Thread rivet nut onto mandrel
  5. 05Fire tool and set fastener
02 / COACH

Guide junior learners live

The AI watches through a beginner's phone or glasses camera. It listens, talks them through the steps in real time, and automatically moves to the next step the exact moment it sees the job done right.

Activity
Rivet nut tool. Sets threaded inserts into sheet metal with no rear access. Three parts: body, mandrel, nose piece.
Do I pick the mandrel first?
Yes. Match it to your rivet nut spec, then torque the nose piece to 15 Nm before attaching.
advance_step({ from: 1, to: 2, method: 'visual' })
03 / TROUBLESHOOT

Diagnose unknown issues

No guide available? Just point the camera at the broken machine and describe the problem. The AI scans your internal library, checks verified manuals online to find a fix, and immediately starts a live coaching session to walk you through the repair.

Describe the broken machine
Machine throws E-04. Compressor short-cycling.
Looks like a refrigerant-side issue. What's on the nameplate?
Carrier 30RAP-080. About 8 years old.
Got it. Searching your library, then the web.
Diagnostic phase
  1. Identify3-phase chiller
  2. DiagnoseLow refrigerant
  3. Fetch3 sources cited
  4. CoachHandoff to learner
Cited sources
  • ↗carrier.com / service / 30RAP / E-04
  • ↗service.refrigeration / short-cycle-diagnosis
  • ↗fieldnotes.io / topics / low-charge-symptoms

One platform. Infinite operational possibilities.

Deploy Retrace wherever your team needs eyes-on-the-job intelligence.

Retrace moves diagnostic data from back-office screens straight to the technician’s eyes. The AI watches the live repair, giving hands-free visual guidance so technicians can fix complex machinery without looking away from the equipment.

Ideal for

Equipment manufacturers, service directors, and technical support leads.

Five elements. One irreplaceable asset.

When you combine first-person sight, spoken intent, and real industrial tasks, you get a highly specialized dataset that cannot be scraped from the web or replicated in a lab.

First-person POV

Same camera the humanoids will use. Ego4D had to convince 855 strangers to wear cameras around. Retrace's data comes from technicians whose existing capture habit is on-glasses or in-phone.

Spoken explanations

Every video is paired with a live, real-time spoken narration explaining exactly why a person is performing an action, connecting visual movement to human intent.

Clear step boundaries

The system records precise markers the exact moment a task step is completed successfully, giving AI models perfect proof of what a job done right looks like.

Learning from mistakes

By capturing both the perfect expert demo and the real-world corrections made by beginners, the data teaches AI how to spot errors and self-correct mid-task.

Industrial verticals

Field service, OEM equipment, skilled trades. Exactly where Ego4D leans thin. The work humanoids will eventually be sold into.

AI copilot for technicians today.
Training corpus for humanoids tomorrow.

Frontier robotics labs do not have a funding problem; they have a data problem. To train humanoid robots to do actual industrial work, they need first-person video of real tradespeople explaining what they are doing. Today, that data is incredibly scarce because most internet video only shows simple household chores.

By helping human technicians complete their work today, Retrace naturally captures the specialized, step-by-step data that the world's most advanced robotics programs are waiting for.

Embodied agent · static
The data-starved labs
Figure
$675M Series B
$2.6B (Feb '24)
OpenAI partnership
1X
$100M Series B
NVIDIA-backed
NEO Beta
Tesla Optimus
Mass-production goal
Public co.
Trained on egocentric video
Physical Intelligence
$400M
$2.4B (Nov '24)
π0 foundation model
Skild AI
$300M
$1.5B (Jul '24)
General-purpose robot brain

Over $1.4B raised in 18 months across 5 frontier humanoid programmes. All on the promise of robots that can do useful physical work.The bottleneck is the same: they are starved for useful data.

Two builders

Full-stack, end-to-end.

Backend / AI + iOS / XR. One real shipped product in four weeks.

  • H
    Hyunseok Hwang
    Backend · AI systems

    FastAPI server, Gemini orchestration (2.5 Pro + 3.1 Live), structured-output extraction, ephemeral tokens, troubleshoot mode, web-grounded search.

  • A
    Arthur Lee
    Design · iOS · XR

    Design across the entire Retrace including SwiftUI app, Ray-Ban HUD design system. Content and storytelling as well.

Frequently Asked Questions (FAQ)

Six questions, six straight answers.

Do I need smart glasses?
No. Retrace runs on iPhone today. The phone holds the direct Gemini Live session, audio and camera. Ray-Ban Meta is an optional capture layer for hands-free work. Glasses and phone are interchangeable per session.
What about the privacy of recorded video?
Expert recordings stay in your tenant. Voice and video stream peer-to-peer between the technician's phone and Gemini Live. The recording is only retained server-side when you publish it as a procedure. Ephemeral tokens scope every Live session to a single voice, tool set, and resumption handle.
Who owns the data?
You do. Procedures, clips, and session telemetry are yours. Egocentric training data is opt-in per-procedure, with named export controls and a per-fleet kill switch.
How is this different from Aquant or Neuron7?
Aquant and Neuron7 are powerful service-intelligence tools you open on a phone or laptop after stopping the work. Retrace runs continuously through the technician's glasses or phone camera while they work: visual auto-advance, voice replies, no swivel-chair.
How is this different from John Deere Service ADVISOR or Cat SIS?
Those are office and shop tools: searchable manuals, diagnostic codes, calibrations. They don't deliver real-time visual guidance on the equipment. Retrace is the layer on top: the technician's hands stay on the machine, the model watches, the procedure auto-advances when the completion criterion is visible.
Deployment model?
Hackathon-stage today. iOS app on TestFlight, backend on AWS Lightsail (migrating to eu-central-1 before Berlin). Production deployment is a self-hosted FastAPI + your Gemini API tenant. No third-party data egress.
AI copilot for technicians today.
Training humanoids tomorrow.

Try Retrace on TestFlight. Or talk to us about industrial pilots.

Install on TestFlight
jacogoby@gmail.comykarthurlee@gmail.com