Beat 01
Autonomy is earned, never assumed
Any vendor promising fab-wide autonomy on day one has not spent a shift on a line.
About
Why we are building display-fab autonomy
Displays are the most-looked-at surface in modern life, and they are made by processes that still lose margin in the gaps between tools. We think that gap is closable.
- Design-partner stage
- Fab-first
- Honest about status
Origin
The thesis
Display manufacturing is one of the most capital-intensive processes on earth, and its economics are decided by uniformity, yield and how fast a new product ramps. All three are still governed by human-paced loops: a drift is noticed at test, a nuisance call queue is worked through by hand, a new model is tuned across quarters by a small number of irreplaceable engineers.
Everything needed to close those loops now exists — panel-level perception, accelerated simulation, robotics control and edge compute that can sit safely beside a tool. What did not exist was a system designed around how a fab actually grants trust: slowly, on evidence, with an audit trail.
So Emiteon is built backwards from the trust model. Shadow first. Advisory second. Control only where the evidence and the twin both say yes. The technology is the easy half.
Beliefs
What we hold to
Beat 02
The override is the product
Engineer corrections are the highest-value data in the building. A system that treats them as friction will never improve.
Beat 03
Process IP is sacred
Fleet learning must travel without recipes, panel designs or defect imagery ever leaving the tenant.
Beat 04
Measure what the fab already measures
If we cannot move a number that is already on a weekly report, we have not done anything.
# closed loop, one panel
perceive substrate → tft → emitter stack → sub-pixel
plan deposition · anneal · transfer · test/bin
act recipe write-back (SECS-GEM)
sense mura ΔL* · sub-pixel opens · particles
optimise yield · ramp · scrap
log immutable audit entry ✓ signed
retrain engineer correction → model
Approach
How we build
Narrow, measurable and honest — in that order.
Step 1 of 6 · wedge
Start at the wedge
One workflow with a number that already hurts.
Step 2 of 6 · shadow
Prove in shadow
No write-back until accuracy is demonstrated.
Step 3 of 6 · console
Build for the engineer
The review console is the product surface that matters most.
Step 4 of 6 · audit
Make it auditable
If a decision cannot be reconstructed, it should not be made.
Step 5 of 6 · expand
Expand on evidence
Tool by tool, module by module, fab by fab.
Step 6 of 6 · retrain
Close the learning loop
Every correction becomes a better model.
What we value
How we work
Fab-first
Decisions get made with process, yield and quality engineers in the room.
Honest claims
Targets are labelled as targets. No invented references.
Deep over broad
We would rather own one workflow completely than touch ten superficially.
Safety by construction
Fail-safe paths and idempotent steps are not optional extras.
Written thinking
Decisions are written down so they can be challenged.
Long horizon
Fab trust is measured in quarters. We build accordingly.
Roadmap
Where we are and where we are going
Phases, not dates. Each phase is gated on evidence from the one before.
- Phase 1
Wedge
Mura and defect perception plus deposition/anneal advisory on design-partner lines.
- Phase 2
Closed loop
Graduated autonomy on low-risk control paths with twin validation.
- Phase 3
Full fab
All seven agents orchestrated across deposition, transfer, handling, test and bin.
- Phase 4
Fleet
Multi-fab governance, custom models and cross-site learning without IP transfer.
Status
Where we actually are
Emiteon is early-stage and at design-partner scale. Company operating status, customers and outcome metrics referenced on this site are aspirational or placeholder until independently verified. We would rather say that plainly than have you find out during diligence.
- Design-partner program open
- Named references gated on written consent
- Outcome numbers labelled as targets
- Happy to be checked
# closed loop, one panel
perceive substrate → tft → emitter stack → sub-pixel
plan deposition · anneal · transfer · test/bin
act recipe write-back (SECS-GEM)
sense mura ΔL* · sub-pixel opens · particles
optimise yield · ramp · scrap
log immutable audit entry ✓ signed
retrain engineer correction → model
Voices from the line
What the fab floor tells us
“We do not lose panels because nobody is watching. We lose them because the signal that mattered was buried under a thousand nuisance calls.”
“Ramp is the whole game. If a new product takes two quarters of manual tuning, that is two quarters of margin we never get back.”
“Transfer yield is our ceiling on microLED. Every dead emitter is a repair cycle or a scrapped backplane.”
Quotes are illustrative composites drawn from discovery interviews with process-integration, yield and quality engineers. Named references are [PLACEHOLDER] pending design-partner consent.
The economics
What the loop is measured on
Emiteon is priced and evaluated on the numbers a fab already reports. These are design-partner targets for the first twelve months of deployment.
$26BDisplay-fab automation, process control, inspection, test and fab software market
~14%Annual growth in the segments Emiteon plays in
135%Net revenue retention target from land-and-expand
99.9%Uptime target for the fab edge runtime
Figures marked as targets are design-partner objectives, not audited results. Company operating status, customers and outcomes are [ASPIRATIONAL] until independently verified.
FAQ
Questions fabs ask first
Through three gates. Shadow mode observes and predicts with no write-back; advisory mode recommends recipe, transfer and binning moves that a process-integration or yield engineer approves; graduated autonomy releases low-risk deposition, anneal and transfer control once measured accuracy and twin validation clear the bar. High-impact decisions stay human-in-the-loop.
Reducing false calls is the wedge, not a side effect. The mura-and-defect agent is trained on true-versus-false-call labels from your own inspection history, so it separates genuine large-area non-uniformity, sub-pixel opens and shorts, particles and stains from nuisance signals, and every call is traceable to the evidence behind it.
Per-tenant isolation with recipes, panel designs and defect images scoped to your tenant, encryption in transit and at rest, SSO/RBAC, an immutable yield/quality-grade audit log and an on-prem or air-gapped option. Fleet learning shares model improvements, never your recipes.
Deposition, photolithography, encapsulation and anneal tools, Mura/AOI inspection and metrology, array and cell test, mass-transfer and repair stations, robotic handling and MES — vendor-neutral via SECS-GEM/HSMS where the tool supports it, with REST and OPC-UA bridges elsewhere.
Every pixel, perfectly uniform.
Come build the loop
Whether as a design partner, a partner OEM or an engineer — the interesting work is on the line.