Typed, ordered, inspectable
A run is a goal, a plan and a graph of typed steps. Every step has an agent, a tool, a duration and a status, and the whole record is kept per lot.
Features
Capability by capability, with the limit on each.
Waferira's features are the mechanics of a system that is allowed to touch production tools: typed runs, a policy engine, named approvals, fused inspection, virtual metrology, a twin, genealogy and an IP-isolated audit log. The list below is design scope, not a shipped feature list.
Illustrative scenario: Module M2 · 300 mm · metal-1 · lot L-48213, one lot through metal-1, one approval gate
Features
What the platform does
Each capability names what it does and where its limit sits.
Recipe bounds enforced at call time
Qualified recipe windows, per-parameter limits, autonomy level, tool state, qualification and interlocks are evaluated before any recipe moves.
Named humans at the gate
Holds, out-of-window moves and lot release wait for the approver the fab names, with the OCAP linked. The request, the reasoning and the decision are logged.
Three modalities, one classifier
Inline optical, e-beam and SEM imagery fused into one defect classification, dispositioned killer or nuisance, localised to wafer, die and lot.
Every wafer scored
Metrology predicted from tool-sensor and FDC signatures on every wafer, with physical metrology routed where the uncertainty is.
Rehearse before you run
Candidate correctables, recipes, WIP and dispatch moves ranked on predicted yield, uniformity, excursion risk and cycle time before anything moves.
The record of every lot
Recipe moves, correctables, FDC signatures, virtual-metrology scores, defect map, approvals and release decision, exportable for excursion and customer review.
Approve, correct, feed back
Process and yield engineers see recommendations with their evidence and citations, approve or correct them, and those corrections train the models.
Append-only, exportable
Every agent action with source signals, retrieved procedures, predicted impact, allowed range, rollback condition, approver and lot scope, isolated per tenant.
Every answer cites its source
Retrieval over recipes, integration specs, SEMI standards and OCAP procedures, tenant-isolated and permission-aware.
The craft, written down
Per-fab yield and process history and per-engineer performance memory, versioned and scoped per tenant, with no cross-tenant leakage.
Many modules, one policy
Model versions promoted by module with canary rollout and automatic fallback; group policy with per-fab override on the Enterprise tier.
Run timeline · scenario
One scenario run, step by step
Every Waferira run is an ordered, inspectable sequence. This is run_7b2e9c on Module M2 · 300 mm · metal-1 · lot L-48213. Scenario completes in 10 steps with one approval gate. Overlay and CD predicted inside spec, one reticle-repeater signature caught and held for review, dispatch re-optimised, lot released with full genealogy.
run_7b2e9c
Module M2 · 300 mm · metal-1 · lot L-48213
illustrative scenario
SUCCEEDED
-
01
ingest.lot
SUCCEEDED
0.7 s
Fab Orchestrator · Pulled lot L-48213 from fab MES: 25 wafers, 300 mm, metal-1 layer, route, reticle, overlay and CD spec and the standing cycle-time target. Locked the target envelope for the run.
-
02
twin.simulate
SUCCEEDED
38 s
Fab Twin · Simulated 64 candidate correctable, recipe and dispatch sets on the as-run fab twin and ranked them on predicted overlay residual, CD uniformity, excursion risk and cycle time.
-
03
litho.correctables
SUCCEEDED
1 m 20 s
Litho-and-Overlay · Per-field overlay correctables and a focus-dose trim written to scanner SCN-07 over SECS/GEM after the previous lot's residual trended toward the control limit. Predicted overlay residual back inside spec.
-
04
etch.uniformity
SUCCEEDED
58 s
Etch-and-Depo · Chamber B on etcher ETCH-12 showed an edge-fast uniformity drift in FDC. Pressure and RF trims written within the qualified recipe window; predicted CD uniformity recovered.
-
05
cmp.removal
SUCCEEDED
46 s
CMP-and-Implant · Downforce and platen-speed trim on CMP-03 against the virtual removal-rate estimate to hold post-CMP thickness and dishing inside the window.
-
06
inspect.classify
SUCCEEDED
continuous
Defect-and-Inspect · Inline optical, e-beam review and SEM imagery fused at the tool edge. 1,412 events classified, most dispositioned nuisance; one reticle-repeater signature on three dies flagged for review.
-
07
vm.score
SUCCEEDED
2.1 s
Metrology-and-Virtual · Virtual metrology scored every wafer at post-etch and post-CMP from tool-sensor and FDC signatures. Two wafers routed to physical metrology on predicted variance; the rest sampled at plan.
-
08
approve.human
APPROVAL
1 m 04 s
Fab Orchestrator · The reticle-repeater signature called for a reticle inspection hold and a recipe move outside the qualified window. Held for the module process engineer and approved at 03:14:27.
-
09
dispatch.route
SUCCEEDED
4.3 s
Yield-and-Dispatch · Dispatch re-optimised across 1,840 lots in WIP under reticle, chamber and qualification constraints. The lot routed to the qualified chamber with the shortest predicted queue; AMHS move issued to the stocker.
-
10
lot.release
SUCCEEDED
1 m 05 s
Quality-and-Conformance · Lot L-48213 moved on: genealogy written (recipe moves, correctables, FDC signatures, VM scores, defect map, approvals) and checked against the layer spec.
Tool calls · scenario
The agent shows its work
Every tool invocation, argument and result is written to an append-only, IP-isolated, human-readable log, and every reasoning step is expandable. Nothing about a run is hidden from the fab.
- 03:02:10mes.read_lot({ lot: "L-48213", layer: "M1", module: "M2" })
- 03:02:11ok 25 wafers · reticle R-M1-114 · overlay and CD spec locked
- 03:02:11twin.simulate({ candidates: 64, horizon: "lot" })
- 03:02:49ok best set #41 · overlay residual 2.6 nm · excursion risk 0.05
- 03:02:50policy.evaluate({ set: 41, autonomy: "L3" })
- 03:02:50ok 7 recipe writes inside qualified window · 1 move requires human approval
- 03:02:51secs.write_recipe({ tool: "SCN-07", correctables: "per_field", dose: "-0.3%" })
- 03:04:11ok predicted overlay residual 3.1 → 2.6 nm · CD on target
- 03:04:12secs.write_recipe({ tool: "ETCH-12", chamber: "B", pressure: "+1.5%", rf: "-0.8%" })
- 03:05:10ok edge-fast drift cleared · predicted uniformity inside window
- 03:05:11secs.write_recipe({ tool: "CMP-03", downforce: "-2%", platen: "+1%" })
- 03:05:57ok virtual removal rate on target · dishing inside window
- 03:05:58vision.stream({ tools: ["INS-02", "EB-01", "SEM-04"] })
- 03:09:40repeater: 3 dies · same field position · reticle R-M1-114 · killer class
- 03:09:41vm.predict({ lot: "L-48213", steps: ["post_etch", "post_cmp"] })
- 03:09:43ok 25 wafers scored · 2 routed to physical metrology
- 03:09:44approval.request({ action: "reticle_hold", recipe_move: "outside_window" })
- 03:14:27ok approved by module process engineer · OCAP linked · logged
- 03:14:28cuopt.dispatch({ wip: 1840, constraints: ["reticle", "chamber", "qual"] })
- 03:14:32ok lot routed to ETCH-12/B · AMHS move to stocker ST-14
- 03:14:33amhs.move({ foup: "F-20911", to: "ST-14" })
- 03:15:21ok carrier delivered · handling exceptions 0
- 03:15:22mes.log_lot({ lot: "L-48213", genealogy: "full" })
- 03:16:27ok scenario complete · 10 steps · 1 hold reviewed · released
Reasoning trace
Plan, thought, action, observation, exactly as the orchestrator would record it.
GoalRun lot L-48213 through metal-1 on module M2, hold overlay and…
Run lot L-48213 through metal-1 on module M2, hold overlay and CD inside the layer spec, catch any excursion before physical metrology, and dispatch on cycle-time target.
PlanSimulate the lot on the twin first. Lead with litho so the cor…
Simulate the lot on the twin first. Lead with litho so the correctables are right before etch, then trim etch and CMP against virtual metrology, keep inspection continuous, and hold anything outside the qualified window for a human.
ThoughtThe previous lot's overlay residual on SCN-07 trended toward t…
The previous lot's overlay residual on SCN-07 trended toward the control limit. On the twin, a per-field correctable update with a small dose trim brings the predicted residual back to 2.6 nm without touching focus.
Actiontwin.simulate(candidates=64) returns set #41: per-field correc…
twin.simulate(candidates=64) returns set #41: per-field correctables and dose minus 0.3% on SCN-07, chamber B pressure and RF trims on ETCH-12, downforce minus 2% on CMP-03.
ObservationSet #41 predicts overlay residual 2.6 nm, CD uniformity inside…
Set #41 predicts overlay residual 2.6 nm, CD uniformity inside window and excursion risk 0.05. Nine candidates cut cycle time further but routed the lot to an unqualified chamber and were discarded.
ActionExecute set #41 under autonomy level L3: seven recipe writes i…
Execute set #41 under autonomy level L3: seven recipe writes inside the qualified window, the reticle hold and the out-of-window move routed to the module process engineer.
ObservationReticle-repeater signature on three dies at 03:09:40 from the …
Reticle-repeater signature on three dies at 03:09:40 from the fused inspection stream. Virtual metrology routed two wafers to physical measurement on predicted variance.
OutcomeScenario complete. Overlay and CD predicted inside spec, the r…
Scenario complete. Overlay and CD predicted inside spec, the repeater held and reviewed with the OCAP linked, the lot dispatched to a qualified chamber and released with full genealogy.
Where it pays
Four loops, four metrics
Waferira is not a dashboard. Each agent closes a loop a fab already measures in yield points, excursions, scrapped lots, review hours and cycle time. The tables show what each loop watches and what it is allowed to move.
Overlay and CD
Correct the scanner before the residual reaches the limit
Overlay residuals and CD drift are predicted per scanner, reticle and layer from metrology and inline defect signals. When the trend heads for the control limit the agent updates correctables and focus-dose inside the qualified window instead of waiting for a rework lot.
- Overlay, CD and focus-exposure response modelled per scanner and layer
- Computational-litho context so a correctable move is mask-aware
- Every correctable and its later yield written into the training set
agent.litho_overlay SUCCEEDED
| Signal | Watches | Moves |
|---|---|---|
| Overlay residual | Predicted per field from metrology trend | Per-field correctables |
| CD | Virtual and sampled metrology | Dose trim inside window |
| Focus | Focus-exposure matrix drift | Focus offset or hold |
Unmeasured wafers
Measure every wafer, virtually
Physical metrology is sampled, so most wafers go unmeasured while uniformity drifts. Vmetra predicts the measurement from tool-sensor and FDC signatures on every wafer, routes the uncertain ones to the metrology tool and hands corrected recipes to the process agents.
- Virtual thickness, uniformity and removal rate on every wafer
- Sampling plan driven by predicted variance, not a fixed rate
- Run-to-run APC fed on every lot-step rather than every sampled one
agent.metrology_vm SUCCEEDED
| Signal | Watches | Moves |
|---|---|---|
| Film thickness | Predicted from FDC per wafer | Deposition time and rate trims via Etch-and-Depo |
| Uniformity | Predicted edge-to-centre profile | Pressure and RF trims inside window |
| Removal rate | Predicted from platen telemetry | Downforce and platen trims via CMP-and-Implant |
Killer or nuisance
Disposition the defect before the engineer opens the image
Inspection tools produce more images than any review team can look at. Scanira fuses optical, e-beam and SEM at the tool edge, classifies each event, auto-dispositions nuisance against killer, and raises a hold on a reticle or pattern excursion while the lot is still in the module.
- Optical, e-beam and SEM fused into one classification
- Reticle repeater and pattern excursion detection with inline holds
- Rare defect classes augmented synthetically and validated against real distributions
agent.defect_inspect SUCCEEDED
| Signal | Watches | Moves |
|---|---|---|
| Defect events | Classified and dispositioned in real time | Hold recommendation, signal to control agents |
| Reticle repeaters | Same field position across dies | Reticle inspection hold via approval |
| Particle excursions | Tool and step correlation | Signal to Clean-and-Diffuse and dispatch |
Cycle time
Rehearse the dispatch, then move the lot
Dispatch runs on rules and WIP bubbles anyway. Dispira re-optimises the dispatch list every few minutes under reticle, chamber and qualification constraints, weighted by live yield risk, and the fab twin checks the plan for cycle-time and yield effect before the AMHS moves a carrier.
- Constrained optimisation across the whole WIP, not one bay
- Hot lots and yield-risk lots weighted explicitly
- Dispatch plan validated on the twin before commit
twin.fabtwin SUCCEEDED
| Signal | Watches | Moves |
|---|---|---|
| WIP by step | Live from MES | Dispatch list to qualified chambers |
| Reticle and chamber state | Tool and qualification status | Routing within constraints |
| Predicted cycle time | Twin simulation of the plan | Plan accepted or re-optimised |
Autonomy
Four levels, set per agent and per tool
A fab does not go from manual APC to unattended control in one step. Waferira makes the level explicit, auditable and reversible at any time. The first release runs shadow mode, then assist mode, then graduated autonomy for low-risk APC, defect and metrology loops.
| Level | What the agent does | What the human does | Where it sits in a pilot |
|---|---|---|---|
| L1 · Advisory | Recommends recipe moves, dispositions and dispatch, with citations | Enters every change manually | Shadow mode, first on the design-partner cell |
| L2 · Supervised | Proposes a recipe write or hold; it executes on approval | Approves each move in the review console | Assist mode, where accuracy and ROI are proven |
| L3 · Bounded | Writes inside the qualified recipe window, tool-state checks and OCAP linkage | Approves holds, out-of-window moves and lot release | Graduated autonomy for low-risk APC, defect and metrology loops |
| L4 · Qualified loops | Runs a qualified loop without prompting, inside its envelope | Sets the envelope; reviews the shift record | Planned; reserved for loops the fab has qualified |
Accelerated computing
Physical AI, at the tool
Computational lithography, defect classification on e-beam and SEM imagery, per-wafer virtual metrology and a full-fab twin are among the most compute-intensive workloads in manufacturing. Waferira keeps IP-sensitive inference and feature storage at the fab edge and runs training, simulation and optimisation on GPUs the fab controls.
Inspection vision
Inline optical, e-beam and SEM imagery and sensor fusion on fab-edge GPUs. 10 to 50 inspection streams per pilot module and sub-200 ms priority defect triage are the design targets.
Virtual metrology and FDC
High-volume multivariate tool-sensor and FDC time-series scoring on every wafer, 1 to 5 seconds per lot-step in the design, with nightly retraining as recipes and chambers drift.
Training
Defect, virtual-metrology and process-response models retrained per process family; 500 to 2,000 GPU-hours per family per quarter is the design target. Dedicated capacity is planned.
Fab twin
As-run process and flow simulation, 1,000 to 10,000 candidate recipe, metrology and dispatch scenarios per fab per day in the design. OVX access is planned.
Synthetic rare events
Rare defect, particle, overlay-drift, CMP-dishing and film-uniformity variants generated and validated against real distributions before promotion. Planned.
Dispatch optimisation
Constrained WIP dispatch re-optimised every 1 to 5 minutes across reticle, chamber, route, tool-state and cycle-time constraints.
AMHS and handling
Local FOUP and robot perception and exception detection at the fab edge. Jetson Thor is planned for higher sensor density.
Process reasoning
Tenant-isolated process Q&A and OCAP reasoning over recipes, SEMI procedures and audit logs, with citations enforced on every answer.
Guardrails
An agent that can move a recipe needs a leash
Waferira writes to production tools over SECS/GEM. Every capability is scoped to a qualified recipe window, every write is policy-checked against tool state, and every action is written to an append-only, IP-isolated audit log the fab owns.
- Recipe bounds. Each agent writes only inside the qualified recipe window for that tool, with explicit per-parameter limits.
- Policy and tool-state checks before every write. Autonomy level, tool state, qualification, interlocks and shift are evaluated before a recipe moves.
- Human-in-the-loop gates. Holds, out-of-window moves and lot release wait for a named approver, with OCAP linkage on every excursion.
- Immutable, IP-isolated audit log. Append-only and exportable. Every action records source signals, retrieved procedures with citations, predicted impact, allowed range, rollback condition, approver and lot scope.
- Rollback. Declared rollback conditions return a recipe to its previous qualified version; loss of the edge node leaves the tool on its own controller and recipe.
- Tenant and IP isolation. Recipes, device designs, defect libraries and process memory never cross a customer boundary. Inference and feature storage stay at the fab edge; an on-prem control plane is available for IP-sensitive fabs.
Compliance posture
| Item | Scope | Status |
|---|---|---|
| SOC 2 Type I and fab-security review | Control plane | QUEUED Planned, first 6 months |
| SOC 2 Type II | Control plane | QUEUED Planned, 6 to 12 months |
| SSO and RBAC | Console and approvals | QUEUED Planned, 6 to 12 months |
| IP-isolation and security assurance package | Fab security and IT and OT review | QUEUED Planned, 6 to 12 months |
| SEMI-standards-cited reasoning | Recipes, integration specs, OCAPs | RUNNING In the design, every answer cites its source |
| Per-tenant IP isolation | Data, models, memory | RUNNING In the design from day one |
| On-prem and fab-edge deployment | IP-sensitive fabs | RUNNING Design option |
FAQ
Straight answers
The questions fab directors and process engineers ask in the first meeting.
Yes, over SECS/GEM, but only inside the qualified recipe window for that tool, with per-parameter limits and tool-state checks, and only at the autonomy level your fab has set. Level 1 is advisory: Waferira recommends and an engineer enters everything. Design partners start there, in shadow mode.
The tool stays on its own controller and its last qualified recipe. Waferira is a supervisory layer on top of the APC, inspection, metrology and MES systems you already run, never a replacement for them, so an outage leaves the module running the way it runs today.
No. Inference and feature storage stay at the fab edge, recipes, device designs, defect libraries and process memory are isolated per tenant, and an on-prem control plane is available for IP-sensitive fabs. Any cross-fab learning is planned, contract-gated and privacy-preserving, and none of it moves recipes or designs.
Advisory recommendations start from the existing APC, metrology and ADC data on the wedge module. Defect classification, virtual metrology and excursion prediction improve as labelled events and lot outcomes accumulate per process family. Rare excursions are augmented synthetically only after validation against real distributions.
You are, the same as with any APC strategy, which is why every write is policy-checked, bounded to the qualified window, logged and reversible, and why holds, out-of-window moves and lot release wait for a named approver with the OCAP linked.
A paid pilot on one process module or tool fleet, scoped to one wedge workflow with a baseline agreed in writing. Shadow mode first, observing and measuring the baseline. Then assist mode, where the agent recommends and your engineer approves and accuracy and ROI are measured. Then bounded autonomy for low-risk APC, defect or metrology loops if the fab is satisfied.
Get started
See the features on one module
Start with one process module or tool fleet and one measurable baseline. A paid, IP-isolated pilot on defect classification, virtual-metrology-driven overlay and CD control or uniformity APC shows the number before you commit further.