Sealix
High-speed fill, seal and label vision that stops leakers and giveaway before they ship.
agent.defect_inspect
Products
Buy the stage that hurts. Add the rest when it pays.
Five products, one line-edge runtime, one audit and genealogy backbone, one data flywheel. Sealix inspects the output and feeds each defect back to the filler, Fermix and Carbyn hold the cellar and the carbonation skid, Pastura closes the safety layer over the loop, and Cuvera simulates the batch before every run.
Illustrative scenario: Plant run · one lager batch from brewhouse to can · brewhouse, cellar, filler, tunnel pasteurizer. Not a measured run.
The ecosystem
Five products, one loop
These are the stages of one brew, ferment, carbonate, fill, pasteurize and inspect loop plus the twin that plans it, not separate bets. They share the line-edge runtime, the sensors and cameras, the food-safety governance, the control connectors and one compounding flywheel.
Fermix
Autonomous fermentation and yeast control that hits attenuation and flavor every batch.
agent.ferment_yeast
Carbyn
Precision carbonation and filtration control that holds clarity and CO2 to exact spec.
agent.filter_carbonate
Pastura
Autonomous pasteurization and CIP assurance that proves every batch is safe and stable.
agent.pasteurize_cip
Cuvera
The as-produced batch digital twin that hits quality, carbonation and takt before the run.
agent.orchestrator
Fill and inspect wedge · agent.defect_inspectSealix
High-speed fill, seal and label vision that stops leakers and giveaway before they ship.
The problem
Beverage lines fill tens of thousands of containers an hour, and a fill-level, seam or cap, label or foreign-body fault becomes loss, giveaway, leakers, recalls and food-safety risk fast. Standalone vision systems flag defects but do not feed the filler to fix the cause, so plants keep over-filling to avoid low fills, which is pure giveaway, and reject downstream instead of correcting upstream. The missed defects are the expensive ones.
What Sealix does
Sealix is Brewira's defect-and-inspect plus adaptive-fill agent. It runs high-speed edge vision for fill-level estimation, seam and cap integrity, label placement and OCR, clarity, container defects and foreign bodies, and closes the loop by feeding drift signals back to fill-and-seal control so the filler corrects toward target level with minimal giveaway. Rare defects are augmented with synthetic data so the models catch faults too infrequent to collect naturally.
Key features
- Fill-level estimation with adaptive filler feedback to cut giveaway and low fills
- Seam, cap and seal integrity and leaker-risk detection
- Label placement, orientation and OCR and date-code checks
- Foreign-body, clarity and container-defect detection
- Synthetic-defect augmentation for rare fault classes
- Sub-100 ms edge reject decisions with an immutable inspection log
AI capabilities
| Capability | Status |
|---|---|
| Fine-tuned high-speed detection, segmentation and OCR vision | design scope |
| Fill-level regression and filler and seamer drift anomaly detection | design scope |
| Cosmos and Replicator synthetic generation of rare defects | design scope |
| Adaptive closed-loop filler correction | bounded per partner; fleet scale planned |
NVIDIA stack
Jetson Orin or Thor at the line running Metropolis, DeepStream and TensorRT; Holoscan to synchronize 8 to 24 camera feeds with analyzer and telemetry streams; Cosmos and Omniverse Replicator for synthetic leakers, low fills, cap skew and foreign bodies; Triton and NIM to serve several vision and anomaly models per line with canary rollout; DGX/HGX or cloud GPUs for fine-tuning and synthetic generation.
- For
- Any high-speed beverage filling or packaging line: beer, soft drinks, water, juice, energy and functional, RTD and dairy-alt, at breweries, beverage makers and co-packers.
- Price
- $14,000 per filling or packaging line per month, expanding to the Plant tier; outcome pricing on measured giveaway and reject reduction.
- First release
- Shadow-mode inspection and fill-level analytics on one line, then assist-mode filler recommendations, then bounded adaptive-fill write-back. One filler or seamer and one camera stack integrated first.
- In the loop
- Consumes carbonation state from Carbyn, feeds fill and seal outcomes to Cuvera and Pastura, and shares cameras, data and governance with every sibling.
Fermentation wedge · agent.ferment_yeastFermix
Autonomous fermentation and yeast control that hits attenuation and flavor every batch.
The problem
Fermentation is the least controllable, highest-variance step in beverage making. Yeast and culture behavior drifts with pitch rate, viability, generation, temperature and gravity, so two identical recipes ferment differently. Brewmasters manage this by hand from spot gravity readings and experience, which caps consistency, stretches tank time and produces off-flavor, over- or under-attenuation and downgraded batches. The craft lives in a few scarce heads and does not scale as SKUs multiply.
What Fermix does
Fermix is Brewira's autonomous fermentation and yeast and culture agent. It fuses inline gravity, dissolved oxygen and CO2, pH, temperature and pressure with a physics-informed fermentation model to predict the attenuation curve, flavor and ester profile and finish time for each tank, then controls temperature ramps, pressure and yeast pitching, cropping and harvest inside brewmaster-approved windows. It learns each strain's and tank's behavior so every batch lands closer to target with less tank time and loss.
Key features
- Per-tank attenuation and flavor-profile forecasting from live sensors
- Yeast and culture viability, pitch-rate, generation and cropping optimization
- Closed-loop temperature and pressure control inside approved recipe windows
- Diacetyl, ester and off-flavor risk prediction with hold and crash timing
- Strain- and tank-specific memory that encodes brewmaster craft
- Immutable action log for every setpoint change, food-safety grounded
AI capabilities
| Capability | Status |
|---|---|
| Physics-informed fermentation kinetics for standard strains | design scope; novel cultures planned |
| Time-series prognostics for attenuation and finish | design scope |
| Anomaly detection for stuck or runaway ferments | design scope |
| Closed-loop self-tuning of curves by strain and tank | planned, validated per partner |
NVIDIA stack
Jetson Orin at the cellar for analyzer fusion and control decisions; CUDA and Modulus for physics-informed fermentation and CFD mixing and heat simulation; TensorRT and Triton for per-tank model serving; RAPIDS for dense time-series feature generation; NeMo and NIM for grounded beverage-process reasoning; DGX/HGX or cloud GPUs for per-strain training (planned).
- For
- Breweries, kombucha, RTD and functional-beverage makers, cideries and beverage groups running fermentation at $100M to $40B revenue.
- Price
- $14,000 per fermentation-line equivalent per month, expanding to the Plant tier; outcome pricing on measured tank-time and downgrade reduction.
- First release
- Shadow-mode attenuation and flavor prediction on one cellar, then assist-mode temperature and pressure recommendations, then bounded write-back on low-risk setpoints. One PLC or SCADA and one analyzer vendor integrated first.
- In the loop
- Feeds attenuation and flavor state to Carbyn and Cuvera, and shares sensors, tenant data and control governance with every sibling.
Carbonation wedge · agent.filter_carbonateCarbyn
Precision carbonation and filtration control that holds clarity and CO2 to exact spec.
The problem
Carbonation and filtration decide mouthfeel, clarity and shelf stability, yet most plants run fixed dosing setpoints and correct after the fact from spot lab checks. CO2 and nitro drift, temperature swings and filter loading push batches out of spec, causing rework, giveaway from over-gassing, flat or over-carbonated product, hazy clarity rejects and downgrades. CO2 is a real cost and, at times, a supply risk, and spending it blindly hurts margin.
What Carbyn does
Carbyn is Brewira's filtration-and-carbonation agent. It fuses inline CO2 and dissolved oxygen, turbidity and clarity, temperature, pressure and flow signals to predict in-line carbonation and clarity, then controls dosing, pressure and filtration inside spec windows to hold target CO2 or nitro and clarity with minimal CO2 waste. It runs shadow, then assist, then bounded autonomy and learns each product's and package's carbonation behavior.
Key features
- Inline carbonation and clarity prediction with closed-loop dosing
- CO2 giveaway minimization and usage and cost tracking per batch
- Filter loading and breakthrough prediction with backwash and change timing
- Product- and package-specific carbonation profiles with memory
- Off-spec and clarity-reject risk alerts with recommended corrections
- Immutable action log for dosing and filtration setpoint changes
AI capabilities
| Capability | Status |
|---|---|
| Inline analyzer fusion and carbonation mass-transfer prediction for standard beverages | design scope |
| Carbonation CFD for novel formats | planned |
| Filter-loading prognostics and anomaly detection | design scope |
| Closed-loop dosing autonomy | bounded, validated per partner |
NVIDIA stack
Jetson Orin at the carbonation and filtration skid; CUDA and Modulus for carbonation mass-transfer and mixing simulation; TensorRT and Triton for versioned serving of fusion and control models; Holoscan for synchronized analyzer and sensor streams; RAPIDS for carbonation and clarity time-series features; NIM and NeMo for grounded reasoning over carbonation and quality specs; DGX/HGX or cloud GPUs for training (planned).
- For
- Soft-drink, sparkling-water, beer, seltzer and hard-seltzer, energy and functional and RTD makers and co-packers running carbonation and filtration at scale.
- Price
- $14,000 per line per month inside the Line tier, expanding to the Plant tier; outcome pricing on measured CO2 and reject reduction.
- First release
- Shadow-mode carbonation and clarity prediction, then assist-mode dosing and filter recommendations, then bounded dosing write-back on one product family. One analyzer and one skid vendor integrated first.
- In the loop
- Takes fermentation state from Fermix, feeds carbonation state to Sealix and Cuvera, and shares sensors, data and governance with every sibling.
Safety and assurance layer · agent.pasteurize_cipPastura
Autonomous pasteurization and CIP assurance that proves every batch is safe and stable.
The problem
Pasteurization and CIP are the last guardrails between a beverage and a recall. Tunnel and flash pasteurizers must deliver exact pasteurization units without over-processing, which wastes energy and hurts flavor, or under-processing, which risks spoilage and food-safety failure. CIP must verify hygiene between runs. Most plants run fixed cycles and spot checks, burning water and energy while carrying spoilage and audit risk, and proving what happened to an auditor is slow and manual.
What Pastura does
Pastura is Brewira's pasteurize-and-CIP agent. It monitors and controls tunnel and flash pasteurization to hit target PU with minimal energy and thermal flavor impact, verifies CIP hygiene from sensor and, where fitted, analyzer signals, predicts spoilage and contamination risk, and produces an assurance-grade, immutable food-safety record for every batch. It runs shadow, then assist, then bounded autonomy under HACCP, FSMA, FSSC 22000 and BRCGS grounding, and is blocked from violating a food-safety control.
Key features
- PU targeting and closed-loop pasteurizer control with energy minimization
- CIP cycle verification and hygiene-risk prediction
- Spoilage and contamination risk scoring per batch
- Auto-generated, immutable food-safety and audit records in the batch genealogy
- Standards-grounded reasoning with citations: HACCP, FSMA, FSSC 22000, BRCGS
- Water and energy tracking and reduction on pasteurization and CIP
AI capabilities
| Capability | Status |
|---|---|
| Thermal and PU prediction and control models | design scope |
| CIP anomaly detection and contamination and spoilage risk scoring | design scope |
| RAG over food-safety standards and plant SOPs with enforced citations | design scope |
| GPU pasteurization heat and CFD simulation for cycle optimization | planned per format |
NVIDIA stack
Jetson Orin at the pasteurizer and CIP skid; CUDA and Modulus for pasteurization heat-transfer and CFD for PU-versus-energy cycle optimization; TensorRT and Triton for thermal and risk models; Holoscan to synchronize pasteurizer and CIP sensor streams; NeMo, Retriever and Guardrails for grounded food-safety reasoning with citations and action schemas; NVIDIA AI Enterprise and NIM for governed, auditable deployment.
- For
- Beer, juice, dairy-alt, RTD and functional-beverage makers running tunnel or flash pasteurization and CIP; quality- and food-safety-driven plants and beverage groups.
- Price
- $14,000 per line per month inside the Line tier, expanding to the Plant tier; outcome pricing on measured energy and water reduction.
- First release
- Shadow-mode PU and CIP monitoring plus audit-record generation on one pasteurizer, then assist-mode cycle recommendations, then bounded control write-back inside food-safety limits.
- In the loop
- Takes fill and seal state from Sealix, contributes safety and genealogy records to the whole plant and to Cuvera, and closes the assurance layer over the loop.
Twin and planning brain · agent.orchestratorCuvera
The as-produced batch digital twin that hits quality, carbonation and takt before the run.
The problem
New SKUs, recipe changes and line trials are expensive and slow. Beverage makers still validate a batch by running it, then fixing what came out wrong, wasting product, tank time, CO2, energy and line hours on trial and error. There is no reliable way to simulate how a specific recipe will actually brew, ferment, carbonate and fill on a specific line before committing raw materials and capacity. Craft knowledge fills the gap, and it does not transfer.
What Cuvera does
Cuvera is Brewira's as-produced batch digital twin. It simulates brewing, fermentation, carbonation and line flow for a given recipe, yeast or culture, package and line, as-produced against as-designed, and auto-optimizes brew and blend, fermentation and carbonation to hit target quality, carbonation and takt before the run. Calibrated against real batch outcomes from the Brewira fleet, it turns new-SKU setup and batch planning from guesswork into a simulated, validated plan handed to Fermix, Carbyn and Sealix for execution.
Key features
- As-produced batch simulation from brew through ferment, carbonate and fill and line flow
- Auto-optimization of brew and blend, fermentation and carbonation to targets
- New-SKU and line-trial planning before committing materials and capacity
- Takt and line-balancing prediction for the specific package and line
- Calibration against real fleet batch outcomes, physics-informed
- Plan hand-off to Fermix, Carbyn and Sealix for execution
AI capabilities
| Capability | Status |
|---|---|
| Physics-informed simulation of fermentation, carbonation, mixing and line flow for standard formats | design scope |
| Calibration against as-produced batch data | design scope; novel SKUs planned |
| Optimization search over recipe and setpoint space | design scope |
| AI-generated line-trial plans | planned, validated per partner |
NVIDIA stack
DGX/HGX or OVX-class GPUs for the Omniverse line twin and CUDA and Modulus fermentation, carbonation, mixing and heat CFD, the heaviest workload in the ecosystem; cuOpt for brew and fill scheduling, changeovers, CIP, water and energy allocation and takt; RAPIDS for calibration over fleet batch data; Omniverse Cloud and NVIDIA AI Enterprise for hosted, governed deployment; NIM to serve plan endpoints per tenant.
- For
- Beverage makers and groups with heavy SKU proliferation and frequent line trials: functional, RTD and seltzer innovators, co-packers and multi-site groups.
- Price
- Included in the Plant tier at $90,000 per month and in Enterprise agreements; premium pricing for innovation-heavy makers on trial-cost and time-to-launch metrics.
- First release
- Simulate one standard product family on one line and calibrate against as-produced batches, then add auto-optimization for brew and carbonation, then plan hand-off to the execution agents.
- In the loop
- Consumes as-produced data from Fermix, Carbyn, Sealix and Pastura and hands optimized plans back to them; the planning brain that ties the siblings together.
Run timeline
The products in one run
Every Brewira run is an ordered, inspectable sequence. This is run_scn_01, an illustrative scenario: Plant run · one lager batch from brewhouse to can · brewhouse, cellar, filler, tunnel pasteurizer. Scenario complete · batch inside spec · one human approval at the pasteurizer.
run_scn_01
scenario · not measured
SUCCEEDED
-
01
order.receive
SUCCEEDED
0.8 s
Plant Orchestrator · Pulled the recipe, the package spec, the HACCP plan with its pasteurization critical control point, the yeast lot record and the standing CIP window from the beverage MES and the lab system; locked the target envelope for the run.
-
02
twin.simulate
SUCCEEDED
38 s
Plant Orchestrator · Simulated 24 candidate brew, fermentation and carbonation plans on the Cuvera as-produced twin; ranked them on gravity margin, predicted finish time, CO2 target, PU margin, takt and CO2 and energy use, and discarded every candidate that thinned the pasteurization margin.
-
03
brew.blend
SUCCEEDED
2 m 40 s
Brew-and-Blend · Held wort gravity to recipe from the inline gravity analyzer and the brewhouse PLC, trimming the blend and utilization setpoints inside the recipe window rather than correcting after a spot lab check.
-
04
ferment.control
SUCCEEDED
5 d 06 h
Ferment-and-Yeast · Predicted the attenuation curve and finish time for the tank from inline gravity, dissolved oxygen and CO2, pH and temperature; held the temperature ramp and timed the crash inside the brewmaster-approved window; no stuck or runaway fermentation was flagged.
-
05
carbonate.dose
SUCCEEDED
3 m 15 s
Filter-and-Carbonate · Fused inline CO2, dissolved oxygen, turbidity, temperature and pressure; held carbonation to the package target with the minimum CO2 and timed the filter backwash from the predicted loading instead of a fixed schedule.
-
06
fill.seal
SUCCEEDED
continuous
Fill-and-Seal · Adjusted fill valves toward the target level inside the approved window using drift feedback from inspection, and ran seam checks on the seamer; the filler corrected upstream instead of rejecting downstream.
-
07
inspect.line
SUCCEEDED
continuous
Defect-and-Inspect · 16 camera streams at the line edge: fill level, seam and cap, label and date code, foreign body and container checks on every can; three low-fill rejects were attributed to one filler valve and Fill-and-Seal corrected it.
-
08
pasteurize.verify
SUCCEEDED
continuous
Pasteurize-and-CIP · Held the tunnel pasteurizer on the validated PU profile at the lowest zone energy inside the envelope; every critical control point reading was written to the audit log against the HACCP plan and the CIP release checks were armed.
-
09
approve.human
APPROVAL
1 m 52 s
Plant Orchestrator · The lowest-energy candidate wanted to move the pasteurizer CIP into a window that overlapped the next brand's start. Held for the shift quality manager, who kept the validated PU profile untouched and approved the CIP move with a tighter release check.
-
10
release.genealogy
SUCCEEDED
1 m 05 s
Quality-and-Conformance · Batch released: gravity and attenuation curves, yeast lot, CO2 and clarity, fill and seal outcomes, inspection rejects, PU readings, CIP release and every human decision written to the batch genealogy and checked against spec.
Compare
What is in each plan
Modules are licensed per line on the Line tier and bundled on Plant and Enterprise.
| Capability | Line | Plant | Enterprise |
|---|---|---|---|
| Lines covered | 1 | All on site | All in group |
| Agent modules | 1 of 11 | All 11 | All 11 + custom models |
| Cuvera twin | Not included | Included | Custom twin formats |
| Line-edge node | Included | Per line | Fleet managed |
| Guarded PLC writes | Scoped tags | All scoped tags | Group policy |
| Batch genealogy | Included | Included | Included + export |
| Autonomy stages | Shadow, assist, bounded | Shadow, assist, bounded | Shadow, assist, bounded |
| Deployment | Line edge + cloud | Plant edge + cloud | On-prem or private cloud |
| Outcome pricing | On request | Available | Standard |
| Indicative price | $14,000 / mo | $90,000 / mo | $650k – $8.5M ACV |
Tool calls
One log, every product
Every tool invocation, argument and result is written to an immutable, human-readable log, and every reasoning step is expandable. Nothing about a run is hidden from the plant. The log below is the scenario run, not a measured one.
- 04:50:02mes.read_order({ batch: "L-0931", brand: "LAGER-33", package: "CAN-330" })
- 04:50:02ok recipe locked · HACCP plan v6 · CCP: tunnel PU 15–20
- 04:50:03lab.read_results({ yeast_lot: "Y-2207", test: "viability" })
- 04:50:03ok viability inside window · lot accepted for pitching
- 04:50:04twin.simulate({ candidates: 24, horizon: "batch" })
- 04:50:42ok best plan #7 · PU margin held · lowest CO2 and energy per hl
- 04:50:43policy.evaluate({ plan: 7, stage: "bounded" })
- 04:50:43ok 7 writes permitted · 1 write requires human approval
- 04:50:44plc.write_setpoint({ tag: "BH1.BLEND_RATIO", basis: "gravity+recipe" })
- 04:53:24ok wort gravity on recipe · utilization held
- 05:10:00ferment.predict_curve({ tank: "FV-12", lot: "Y-2207" })
- 05:10:00ok attenuation curve predicted · finish day 5 · crash timed
- 05:10:01plc.write_setpoint({ tag: "FV12.TEMP_RAMP", window: "brewmaster_approved" })
- 05:10:01ok ramp applied · no stuck or runaway ferment flagged
- 06:02:10plc.write_setpoint({ tag: "CARB1.CO2_DOSE", basis: "co2+do+turbidity" })
- 06:05:25ok CO2 on package target · clarity inside spec · backwash timed
- 06:05:26vision.stream({ cameras: 16, models: ["fill_level", "seam_cap", "label_ocr", "foreign_body"] })
- 06:31:48reject: low fill · can 20417 · attributed to filler valve 14 drift
- 06:31:49defect.attribute({ event: "low_fill", cause: "filler.valve_14" })
- 06:31:49ok fill_seal notified · valve 14 trimmed inside window · two more rejects, then clean
- 06:31:50plc.verify_pu({ tag: "TUN1.ZONES", profile: "validated" })
- 06:31:50ok PU inside profile · readings → audit log
- 06:31:52approval.request({ action: "cip.window", reason: "overlaps next brand start" })
- 06:33:44ok approved by the shift quality manager · PU profile untouched · tighter release check
- 09:12:30mes.release_batch({ batch: "L-0931", genealogy: "full" })
- 09:13:35ok batch released · scenario complete
Reasoning trace
Goal, plan, thought, action, observation, exactly as the orchestrator would record it.
GoalBrew batch L-0931 to recipe gravity, ferment it to target atte…
Brew batch L-0931 to recipe gravity, ferment it to target attenuation, carbonate and fill it to spec at line speed, pasteurize it on the validated PU profile and release it with full genealogy.
PlanSimulate the whole batch on the twin first. Lead with gravity …
Simulate the whole batch on the twin first. Lead with gravity so the cellar gets the right wort and the filler gets the right CO2. Verify the pasteurization critical control point continuously. Let the filler chase target level only inside the approved window; hold anything beyond the envelope for a human.
ThoughtThe standing practice over-fills to avoid low fills and over-g…
The standing practice over-fills to avoid low fills and over-gasses to avoid flat product. With continuous fill-level estimation and inline CO2 the filler and the carbonator can sit at target. The saving is giveaway, not safety margin: the pasteurization CCP is never a candidate for optimization.
Actiontwin.simulate(candidates=24) → plan #7: gravity on recipe, fin…
twin.simulate(candidates=24) → plan #7: gravity on recipe, finish day 5, CO2 at package target, PU margin unchanged, lowest predicted CO2 and energy per hectoliter.
ObservationPlan #7 wants the pasteurizer CIP in a window that overlaps th…
Plan #7 wants the pasteurizer CIP in a window that overlaps the next brand's start. The policy engine routes that single move to the shift quality manager and permits the other seven writes.
ActionExecute plan #7 at the bounded stage: brewhouse blend, tank ra…
Execute plan #7 at the bounded stage: brewhouse blend, tank ramp, CO2 dosing and fill-valve writes proceed; PU is verified continuously; the CIP move waits.
ObservationApproval granted at 06:33:44 with the PU profile untouched and…
Approval granted at 06:33:44 with the PU profile untouched and a tighter CIP release check. Three low-fill rejects were attributed to filler valve 14 and corrected by Fill-and-Seal inside the approved window.
OutcomeScenario complete. The batch is released with gravity and atte…
Scenario complete. The batch is released with gravity and attenuation curves, yeast lot, CO2 and clarity, fill and seal outcomes, PU readings, CIP release and the human decision in the genealogy. No measured plant result is claimed; this is the run the platform is designed to execute.
Autonomy
Four stages, set per loop
A plant does not go from manual to bounded control in one step. Brewira makes the stage explicit per loop, auditable and reversible at any time, and some decisions never leave the person.
| Stage | What the agent does | What the person does | When |
|---|---|---|---|
| Shadow | Predicts gravity, attenuation, carbonation, fill and seal faults, contamination and takt; writes nothing | Runs the plant as today; compares predictions to lab, QA and PLC outcomes | First phase with every design partner |
| Assist | Recommends bounded brew, fermentation, carbonation, fill and seal, pasteurization and CIP moves with the evidence | A brewmaster, process engineer or quality manager approves or rejects each one | After accuracy is proven against the baseline |
| Bounded | Writes low-risk filling, carbonation and inspection setpoints inside documented recipe, quality, food-safety, equipment and rollback limits | Approves anything above the limits; reviews the shift record | After the food-safety and ROI gates agreed up front |
| Exceptions | Escalates anything outside its envelope, CCP-adjacent moves and release decisions | Always decides these; the agent never does | Permanent |
Accelerated computing
Physical AI, at the plant edge
Lines fill tens of thousands of containers an hour, so perception has to be local and fast. Brewira runs GPU inference at the line edge and keeps training, simulation and optimization in the cloud, in a private cloud or on-prem. Items marked planned are on the roadmap, not running.
Edge perception
8 to 24 camera feeds per line through DeepStream pipelines and TensorRT detectors, segmentation, OCR and fill-level models; Holoscan synchronizes cameras with gravity, CO2 and clarity analyzers and filler and pasteurizer telemetry. Design target: sub-100 ms reject decisions.
Model serving
Vision, time-series, anomaly, process and food-safety reasoning models behind one orchestrator, with version routing by brand, package and line. Design target: 20 to 60 model versions per enterprise tenant.
Training (planned)
Line vision, time-series forecasters, anomaly models and NeMo-based beverage reasoning fine-tuned from labeled batch outcomes, brewmaster corrections, lab results and release decisions. Design target: weekly retraining for wedge models, monthly for validated process models.
Digital twin (planned)
The Cuvera as-produced line twin: CUDA-accelerated fermentation, carbonation, mixing and pasteurization heat simulation run before the batch, for new SKU setup and batch planning when a trial is expensive.
Synthetic data (planned)
Rare leaker, low-fill, overfill, cap-skew, seam, label-shift, condensation, glare and foreign-body scenarios generated with domain randomization and validated against real plant events before use.
Optimization
Brew and fill schedules, changeovers, CIP cycles, water and energy allocation and takt balancing under line, tank, product and sanitation constraints, recomputed on the GPU when the plant is disrupted.
Data engineering
Image, time-series and batch-genealogy ETL for feature generation, quality correlation and cross-plant benchmarks on the GPU.
Robotic handling
Pack, case and palletizing handling, with perception and motion policies built and validated in simulation before they move a pallet.
Design targets
Product design targets
The numbers Brewira is engineered against, from the GPU utilization plan.
<100 ms
edge inference for a reject decision at line speed
<250 ms
drift alert to the fill-and-seal agent
8–24
camera feeds per high-speed filling or packaging line
60–240 FPS
aggregate frames processed per line, depending on container speed
20–60
deployed model versions per enterprise tenant, with canary rollout and rollback
100k–1M
labeled synthetic rare-defect variants per design-partner phase
Design targets from the GPU utilization plan, all marked aspirational in the architecture. Nothing on this page is a measured plant result; measurements start when the first design-partner line runs in shadow mode.
From the plant floor
Who this is built for
The three people in a beverage plant whose week changes when the process runs itself, and what each of them needs from Brewira.
Plant or Operations Director
Owns production rate, quality, loss, giveaway and yield, water and energy, and food safety. Wants a plant that holds spec at higher rate with fewer interventions, and a number that justifies the spend.
Brewmaster or Process Engineer
Owns recipes, brewhouse and blend, fermentation and carbonation process windows and yield. Approves and corrects the agents in assist mode; those corrections train the models.
Quality or Food-Safety Manager
Owns product quality and safety under HACCP, FSMA, FSSC 22000 and BRCGS, contamination and fill and seal conformance, and batch genealogy. Needs every action grounded, cited and written to an immutable record.
IT/OT and sustainability leads approve the deployment. There are no customer quotes on this site yet; design partners will speak for themselves when they choose to.
Integrations
It speaks plant, not cloud
Brewira reads and writes through the systems already on the floor. No parallel historian, no new HMI, no rip-and-replace. Connectors are built against each design partner's stack first; long-tail integrations are out of scope for the first release.
PLC and SCADA of brewhouse, blending and syrup systems, fermentation tanks and yeast handling
Setpoint reads, guarded writes, interlocks untouched
Filtration and carbonation skids and their inline CO2, dissolved-oxygen and turbidity analyzers
Continuous carbonation and clarity state
High-speed filler, seamer and capper controls
Telemetry in; fill-valve and seam settings inside approved windows
Tunnel and flash pasteurizers, CIP skids and hygiene sensors
PU readings verified and logged; CIP release checks
Multi-camera stations for fill level, seal and cap, label, foreign body and container defects
Frames, defect events, reject commands
Orders, recipes, product specs, batch genealogy, LIMS results
Targets in, release records out
Plant and asset meters; Isaac-driven pack, case and palletizing cells
Hard constraints for scheduling; bounded handling
SSO and role-based access; SOC 2 on the roadmap
Named approvers, scoped roles
Guardrails
An agent that can move a setpoint needs a leash
Brewira writes to production equipment in a food plant. Every capability is scoped, every write is policy-checked, every output is cited, and every action is written to an append-only audit log the plant owns.
- Bounded action space. Each agent can only write to an explicit tag allow-list, inside per-tag rate and magnitude limits, at the autonomy stage the plant has set.
- Critical control points are read, never written. Pasteurization units, CIP release checks and batch release decisions stay with the HACCP plan and the people who own it.
- Policy engine before every write. Autonomy stage, interlock state, product, shift and operator presence are evaluated before a setpoint moves.
- Grounded and cited. Every recommendation carries a citation to the recipe, spec or procedure it relied on. Retrieval is tenant-isolated and permission-aware.
- Immutable audit log. Append-only, exportable and retained on the plant's own storage, aligned to HACCP, FSMA, FSSC 22000 and BRCGS record-keeping.
- Hard fallback. If the edge node, the network or a model stops, the plant keeps running on its existing PLC and SCADA control and interlocks.
- Tenant and IP isolation. Recipes, supplier data and tolerance models never cross a customer boundary. On-prem deployment available.
Assurance posture
| Item | Scope | Status |
|---|---|---|
| SOC 2 Type I | Cloud control plane | QUEUED Roadmap, first 6 months |
| SOC 2 Type II | Cloud control plane | QUEUED Roadmap, 6 to 12 months |
| HACCP · FSMA · FSSC 22000 · BRCGS | Grounding and record-keeping | RUNNING Design basis for retrieval and the audit log |
| Per-tenant isolation | Data, memory, models | RUNNING Design rule |
| On-prem deployment | Recipe-sensitive producers | RUNNING Design option |
Brewira supports your food-safety records. It is not a certification and it does not replace your HACCP plan.
Enterprise
Standardize autonomy across the group
One policy model, one audit trail and one set of yield, loss, giveaway, water and energy benchmarks across every plant in the group, with each site's recipes and tolerance models kept private to that site.
- Multi-site plant-edge fleet with shared model governance
- Custom recipe, product and tolerance models per site
- SSO and role-based access, named approvers (roadmap, 6 to 12 months)
- Model versioning, validation reports and rollback
- On-prem or private-cloud deployment
- SLAs and outcome-based components on the highest-value workflows
FAQ
Straight answers
The questions plant directors, brewmasters and quality managers ask in the first meeting.
Only at the autonomy stage your plant has set, and only to an explicit allow-list of low-risk loops inside documented limits. Every design partner starts in shadow mode, where Brewira predicts and writes nothing. Assist mode follows, where a brewmaster or process engineer approves each recommendation. Bounded autonomy on filling, carbonation and inspection loops comes after the food-safety and ROI gates agreed up front.
Critical control points are not optimization variables. The pasteurization PU profile, the CIP release checks and the batch release decision stay with your HACCP plan and your people. Brewira verifies readings, writes them to an immutable audit log and attaches them to the batch genealogy. It is grounded in HACCP, FSMA, FSSC 22000 and BRCGS procedures; it is not a certification and does not replace your plan.
Brewira sits above your existing automation as a supervisory layer. If it stops, the plant keeps running on its current PLC and SCADA control and the interlocks it already has. A Brewira outage degrades the plant to the way it runs today, not to a stop.
The first phase is shadow mode on one line: Brewira reads your analyzers, PLC history, lab results and MES while you run the plant as usual, and we measure its predictions against your lab, QA and PLC outcomes. Recommendations start when accuracy is proven against that baseline, and improve as plant-specific history and brewmaster corrections accumulate.
Recipes, product specs, tolerance models and supplier data are tenant-scoped and never used to train another customer's models. Retrieval is multi-tenant isolated and permission-aware. Cross-plant learning, where a customer opts in, is federated and IP-isolated. On-prem deployment is available for producers protecting sensitive recipes.
You are, the same as with any control strategy. That is why every write is policy-checked, bounded, logged and reversible, every output carries a citation to the recipe or procedure it relied on, and anything above your threshold waits for a named approver. The audit log records the request, the reasoning, the limits applied and the human decision.
As a design partner, on one line, with one hard metric: fill giveaway and low-fill rejects, leaker and seal faults, attenuation variance and tank time, CO2 use and clarity rejects, or water and energy on pasteurization and CIP. We instrument the line in shadow mode, measure the baseline, move to assist mode when accuracy is proven, and expand only when the number holds.
Full multi-module autonomy across the whole plant, long-tail integrations beyond the design partners' stacks, and self-serve onboarding. The first release is one wedge workflow, its connectors, the agent loop with human checkpoints, the review console and the assurance-grade food-safety audit log.
Get started
Pick your first product
Most plants start with Sealix on a filler or Fermix on a cellar. Tell us which number you need to move.