Products

Buy the stage that hurts. Add the rest when it pays.

Five products, one plant-edge runtime, one audit and genealogy backbone, one data flywheel. Synthex runs the front-end chemistry, Granulex forms the solids, Granvis is the shared perception layer that feeds every agent, Blendex blends, bags and handles, and Nitrotwin simulates the complex before every run.

Five productsOne runtime$22,000 per unit per month

run_scn_01 · agent graph · scenario 10 nodes · 1 approval gate
01 · orchestrator order.receive ✓ SUCCEEDED 02 · orchestrator twin.simulate ✓ SUCCEEDED 03 · reform_synth loop.control ✓ SUCCEEDED 04 · react_evap evaporate.control ✓ SUCCEEDED 05 · granulate granulate.control ✓ SUCCEEDED 06 · coat_condition condition.cool ✓ SUCCEEDED 07 · defect_inspect inspect.granules ✓ SUCCEEDED 08 · orchestrator approve.human ◆ APPROVAL 09 · blend_bag blend.bag ✓ SUCCEEDED 10 · quality_conf release.genealogy ✓ SUCCEEDED

Illustrative scenario: Complex run · one urea granulation campaign from synthesis gas to bagged lot · ammonia loop, urea synthesis and evaporation, granulator, dryer and cooler, NPK bagging line. Not a measured run.

The ecosystem

Five products, one loop

These are the stages of one synthesize, evaporate, granulate, condition, blend and inspect loop plus the twin that plans it, not separate bets. They share the plant-edge runtime, the cameras and analyzers, the safety governance, the DCS, APC and MES connectors and one compounding flywheel.

Front-end chemistry wedge

Synthex

Autonomous ammonia, urea and nitrate synthesis at exact conversion and energy.

agent.reform_synth

Solids-forming wedge

Granulex

Autonomous granulation, prilling and anti-caking to exact granule spec.

agent.granulate

Shared perception layer

Granvis

See every granule, nutrient and emission fault before it becomes off-spec.

agent.defect_inspect

Blend, bag and handling wedge

Blendex

Autonomous NPK blending, bagging and robotic handling to exact nutrient ratio.

agent.blend_bag

Twin and planning brain

Nitrotwin

Hit nutrient spec, granule quality, energy, emissions and takt before the run.

agent.orchestrator
Front-end chemistry wedge · agent.reform_synth

Synthex

Autonomous ammonia, urea and nitrate synthesis at exact conversion and energy.

The problem

Ammonia synthesis, urea and nitrate reaction and evaporation are the most energy- and emissions-heavy steps in the fertilizer complex, run on fixed setpoints and craft. A reforming or ammonia-loop conversion drift, a urea or nitrate concentration error or an evaporation fault burns enormous energy and CO2 and NOx, and pushes product off-spec before it ever reaches granulation. Ammonia is also safety-critical. Plants run blind on real-time conversion, concentration and energy across the front end.

What Synthex does

Synthex is the reform-and-synthesize plus react-and-evaporate agent for the fertilizer front end. It controls reforming, the high-pressure ammonia loop, conversion and temperature, then urea or nitrate reaction, concentration and evaporation, holding exact conversion and concentration at minimum energy and emissions inside safety-gated action envelopes. It validates candidate loop, reaction and evaporation moves on the Nitrotwin before applying them, and hands a specified melt to Granulex downstream.

Key features

  • Closed-loop reforming and ammonia-loop conversion and temperature control
  • Urea and nitrate reaction and evaporation concentration control
  • Energy per tonne and CO2, NOx and ammonia-slip minimization under conversion targets
  • Safety-gated autonomy with human-in-the-loop checkpoints for ammonia
  • Twin-validated setpoint moves and lot-genealogy-linked energy and emissions logging

AI capabilities

Synthex AI capabilities and status
CapabilityStatus
Reforming, ammonia-loop and reaction-and-evaporation models with compressor and reactor prognosticsdesign scope
Process and CFD surrogate coupling to the twinplanned; fidelity targets aspirational
Process-engineering and emissions-reasoning model, citeddesign scope
Closed-loop, safety-gated synthesis controlbounded per partner after the gates

NVIDIA stack

H100 or HGX to train synthesis and evaporation models and accelerate ammonia-loop and CFD surrogates; Jetson Orin or Thor at the plant edge for deterministic loop control inference; DGX or OVX for twin-coupled process simulation and fleet retraining; CUDA and Modulus for physics-ML surrogates of the ammonia loop and evaporation train; TensorRT for deterministic-latency edge models; Triton to serve synthesis and prognostic endpoints; cuOpt for energy and emissions allocation across the front end; Omniverse as the process twin Synthex validates against; NVIDIA AI Enterprise and NIM for governed, regulated serving.

For
Nitrogen producers of ammonia, urea and nitrate and NPK complexes; reforming, ammonia and urea or nitrate synthesis licensors; advanced-process-control and DCS OEMs; decarbonizing and green-ammonia operators.
Price
$22,000 per synthesis unit per month, expanding to the Complex tier; outcome pricing on measured energy and emissions savings.
First release
Advisory conversion and energy optimization on one ammonia or urea line using existing DCS and analyzer data, then closed-loop, safety-gated synthesis and evaporation control with twin validation.
In the loop
Consumes Granvis sensing, validates in Nitrotwin and hands a specified melt to Granulex; the upstream synthesis brain of the shared loop.
Solids-forming wedge · agent.granulate

Granulex

Autonomous granulation, prilling and anti-caking to exact granule spec.

The problem

Granulation and prilling decide granule size, roundness, hardness and moisture, and a size, hardness or moisture fault drives dust loss, off-spec product, recycle and caking in the bag. Coating and conditioning are craft-bound and run on fixed recycle and curves, so plants overspend on recycle and energy and still ship out-of-spec, dusty or caking granules that customers reject.

What Granulex does

Granulex is the granulate-and-prill plus coat-and-condition agent. It controls granulation or prilling, recycle ratio and bed conditions to hit target granule size, roundness, hardness and moisture, then controls coating and anti-caking conditioning, using Granvis granule vision and analyzer state as inputs and validating recycle, spray and temperature moves on the Nitrotwin before applying them. It minimizes dust, recycle and off-spec while holding the granule spec.

Key features

  • Closed-loop granulation or prilling, recycle and moisture control to granule spec
  • Coating and anti-caking conditioning control
  • Dust and recycle minimization under size and hardness constraints
  • Twin-validated recycle, spray and temperature moves, as-run against as-designed
  • Lot-genealogy-linked granule-quality and energy logging

AI capabilities

Granulex AI capabilities and status
CapabilityStatus
Granulation and prilling recycle, moisture and hardness models with granulator and drum prognosticsdesign scope
GPU-accelerated granulation and CFD surrogate coupling to the twinplanned; fidelity targets aspirational
Process-reasoning model for recipe guidance, citeddesign scope
Closed-loop recycle, spray and conditioning controlbounded per partner after the gates

NVIDIA stack

H100 or HGX to train granulation, moisture and hardness models and granulation surrogates; Jetson Orin or Thor at the plant edge for real-time recycle and spray control inference; DGX or OVX for twin-coupled granulation simulation and fleet retraining; CUDA and Modulus for physics-ML granulation, recycle and drying surrogates; TensorRT for deterministic-latency edge control models; Triton to serve granulation and prognostic endpoints; cuOpt for recycle-loop and throughput balancing under constraints; Omniverse as the granulation twin Granulex validates against; NVIDIA AI Enterprise and NIM for governed serving.

For
Nitrogen, phosphate, potash and NPK granulation and prilling plants; granulation-control and coating-system OEMs; producers fighting dust, caking and product returns.
Price
$22,000 per granulation unit per month, expanding to the Complex tier; outcome pricing on measured dust, recycle and off-spec reduction.
First release
Advisory granule-size and moisture optimization on one granulation line using existing vision and analyzer data, then closed-loop recycle, spray and conditioning control with twin validation.
In the loop
Takes Synthex's melt, uses Granvis inputs, validates in Nitrotwin and hands specified granules to Blendex; the solids-forming brain of the shared loop.
Shared perception layer · agent.defect_inspect

Granvis

See every granule, nutrient and emission fault before it becomes off-spec.

The problem

Fertilizer quality, meaning nutrient content, granule size and shape, dust, caking and moisture, is still checked by periodic lab assays and manual granule inspection, so faults are caught tonnes too late. Emissions of CO2, NOx and ammonia slip are measured sparsely. Without continuous, control-grade perception the synthesis, granulation and blending agents run half-blind, and off-spec product, dust loss and emissions excursions slip through.

What Granvis does

Granvis is the defect-and-inspect plus quality-and-nutrient sensing agent. It fuses granule vision for size, shape, dust, caking and broken prills with online nutrient and moisture analyzers and emissions sensing into a continuous, control-grade quality signal, and predicts off-spec, caking and emissions faults before they happen. It is the shared perception layer that feeds Synthex, Granulex and Blendex, and its detectors are stress-tested against Nitrotwin scenarios.

Key features

  • Granule size, shape, dust, caking and broken-prill vision at line speed
  • Online nutrient-content and moisture analysis fused with vision
  • CO2, NOx and ammonia-slip emissions sensing and excursion prediction
  • Off-spec and caking prediction seconds to minutes ahead
  • Assurance-grade quality and emissions audit trail tied to lot genealogy

AI capabilities

Granvis AI capabilities and status
CapabilityStatus
Fine-tuned granule-vision and analyzer models plus emissions modelsdesign scope
Multimodal sensor fusion at 1 to 10 Hz for control-grade contextdesign scope; fusion-rate targets aspirational
Synthetic rare-fault augmentation with Cosmos and Replicatorplanned
Closed-loop feeds to the control agentsafter the shadow and assist gates

NVIDIA stack

Jetson Orin or Thor at the plant edge across synthesis, granulation and bagging cameras; H100 or HGX to train granule-vision, analyzer and emissions models; IGX and Holoscan dev kits for sensor-fusion pipelines; Metropolis, DeepStream and TensorRT for multi-camera granule, dust and caking vision; Holoscan to fuse vision, analyzer, historian, emissions, acoustic and thermal streams; Cosmos and Omniverse Replicator for synthetic rare off-spec, caking and upset scenarios; Triton to serve vision, analyzer and emissions endpoints; NVIDIA AI Enterprise and NIM for governed, regulated serving; Omniverse as the scenario source that stress-tests the detectors.

For
Nitrogen, phosphate, potash and NPK producers; granule and nutrient online-analyzer OEMs; quality and HSE teams under tightening emissions limits and customer spec audits.
Price
$22,000 per unit per month for granule-sensing AI, expanding to the Complex tier; enterprise custom nutrient and emissions models and edge camera fleets.
First release
Continuous granule-vision plus nutrient and moisture inspection on one line with off-spec alerts, then fused emissions prediction and closed-loop feeds to the control agents.
In the loop
The shared perception feed for Synthex, Granulex and Blendex, stress-tested in Nitrotwin; the eyes of the shared loop.
Blend, bag and handling wedge · agent.blend_bag

Blendex

Autonomous NPK blending, bagging and robotic handling to exact nutrient ratio.

The problem

NPK blending must hit an exact nutrient ratio without segregation, then bag and load to exact weight and spec, while heavy, repetitive bag and bulk handling still leans on a scarce, aging workforce. A blend-ratio error or segregation ruins nutrient spec; a bagging or loading fault drives rework, claims and downtime. Plants run these lines on fixed setpoints and manual handling, blind to real-time ratio and segregation.

What Blendex does

Blendex is the blend-and-bag plus robot-and-handling agent. It controls NPK blend ratio and segregation and bagging and loading to exact nutrient spec and weight, using Granvis nutrient sensing as feedback and validating blend and sequencing moves on the Nitrotwin. It also drives robotic bag, bulk and materials handling with Isaac and Jetson, closing the physical loop from blended product to loaded truck.

Key features

  • Closed-loop NPK blend-ratio and anti-segregation control to nutrient spec
  • Bagging and loading weight and spec control with rework minimization
  • Robotic bag, bulk and materials handling and palletizing
  • Twin-validated blend and line-sequencing moves
  • Lot-genealogy-linked nutrient-ratio and shipment traceability

AI capabilities

Blendex AI capabilities and status
CapabilityStatus
NPK blend-ratio and segregation models plus bagging and handling vision and robot policiesdesign scope
Isaac-trained manipulation policies with synthetic augmentationplanned; sim-to-real targets aspirational
Process-reasoning model for blend recipe guidance, citeddesign scope
Closed-loop anti-segregation control and robotic handlingbounded per partner after the gates

NVIDIA stack

Jetson Orin or Thor for on-robot and edge inference for handling and blend control; H100 or HGX to train blend, segregation and manipulation models; DGX or OVX for Isaac Sim policy training and twin-coupled line simulation; Isaac Sim, ROS and Lab for robotic bag and bulk handling and palletizing policies; TensorRT for deterministic-latency edge and on-robot models; cuOpt for bagging, loading and line-sequencing optimization; Triton to serve blend, vision and robot-policy endpoints; Omniverse as the bagging and handling twin shared with Nitrotwin; NVIDIA AI Enterprise, NIM and Fleet Command for governed serving and edge robot fleets.

For
NPK blenders and bulk-blend plants; bagging and palletizing OEMs; bulk-materials-handling integrators; producers facing labor shortages on bagging and loading lines.
Price
$22,000 per blend or bagging unit per month, expanding to the Complex tier; robotic handling as a premium enterprise add-on; outcome pricing on rework and claim reduction.
First release
Advisory NPK blend-ratio and bagging control on one line using existing sensors, then closed-loop anti-segregation control and Isaac-driven robotic handling.
In the loop
Takes Granulex's granules, uses Granvis nutrient feedback, validates in Nitrotwin and closes the physical loop to shipment; the blend, bag and handling brain of the shared loop.
Twin and planning brain · agent.orchestrator

Nitrotwin

Hit nutrient spec, granule quality, energy, emissions and takt before the run.

The problem

A fertilizer complex is a giant, drifting, energy- and emissions-hungry system, yet producers change synthesis, granulation and blend recipes and sequence products on operator craft and fixed procedures. There is no way to test a product change, a grade changeover or an energy and emissions trade-off before committing the plant, so changeovers waste tonnes, energy and CO2 and NOx, and off-spec risk is discovered live.

What Nitrotwin does

Nitrotwin is the as-produced fertilizer-complex digital twin and yield-and-takt optimizer. It simulates the ammonia loop, urea or nitrate reaction and evaporation, granulation or prilling and NPK blending together to hit target nutrient spec, granule quality, energy, emissions and takt before the run. It validates every Synthex, Granulex and Blendex move, sequences product and grade changeovers, and optimizes yield, off-spec and recycle, energy and emissions across the whole complex.

Key features

  • Coupled complex twin: ammonia loop, evaporation, granulation and blending
  • Pre-run validation of every control-agent move, as-produced against as-designed
  • Product and grade changeover sequencing and takt balancing
  • Whole-complex yield, off-spec and recycle, energy and emissions optimization
  • Scenario library that stress-tests recipes and Granvis detectors

AI capabilities

Nitrotwin AI capabilities and status
CapabilityStatus
Physics-informed process surrogates for the ammonia loop, evaporation, granulation, blending and emissionsdesign scope; fidelity targets aspirational
GPU-accelerated process and CFD simulation sweepsplanned
Constrained optimization over yield, energy, emissions and taktdesign scope
Process-reasoning model, citeddesign scope

NVIDIA stack

DGX or OVX for the twin's coupled process and CFD simulation and scenario sweeps; H100, HGX or GB200 to train surrogates and run large simulation batches; Jetson Orin or Thor for edge execution of twin-approved run plans on the agents; Omniverse as the complex twin environment and for scenario authoring; CUDA and Modulus for physics-ML surrogates of the ammonia loop, evaporation and granulation; cuOpt for yield, energy, emissions, takt and changeover-sequence optimization; Cosmos and Omniverse Replicator for synthetic upset and changeover scenarios; NVIDIA AI Enterprise and NIM for governed serving; Omniverse Cloud and Base Command for twin simulation and fleet training at scale.

For
Multi-unit fertilizer complexes and multi-site producers; process-technology licensors; plant-optimization and MES teams; sustainability and energy leaders managing decarbonization.
Price
Core of the Complex tier at $140,000 per month and of Enterprise agreements from $900k to $12M ACV; outcome upside on changeover, energy and emissions savings.
First release
A surrogate-based twin of one unit, the ammonia loop or a granulator, validating agent moves and one changeover, then the coupled complex twin with whole-plant optimization.
In the loop
Simulates and sequences Synthex, Granulex and Blendex and stress-tests Granvis; the planning and validation brain that ties the whole loop together.
Run timeline

The products in one run

Every Nitreum run is an ordered, inspectable sequence. This is run_scn_01, an illustrative scenario: Complex run · one urea granulation campaign from synthesis gas to bagged lot · ammonia loop, urea synthesis and evaporation, granulator, dryer and cooler, NPK bagging line. Scenario complete · lot inside spec · one human approval on the ammonia loop.

run_scn_01 scenario · not measured SUCCEEDED
  1. 01 order.receive SUCCEEDED 0.8 s

    Complex Orchestrator · Pulled the grade spec, the nutrient target, the granule spec for size distribution, hardness and moisture, the emissions permit limits, the safety-instrumented envelope and the standing changeover window from the plant MES and the historian; locked the target envelope for the campaign.

  2. 02 twin.simulate SUCCEEDED 41 s

    Complex Orchestrator · Simulated 24 candidate loop, evaporation, granulation and sequencing plans on the Nitrotwin as-produced complex twin; ranked them on conversion, energy per tonne, predicted granule distribution, recycle ratio, emissions margin and takt, and discarded every candidate that thinned the safety or permit margin.

  3. 03 loop.control SUCCEEDED continuous

    Reform-and-Synthesize · Held ammonia-loop conversion and reformer temperature inside the approved envelope from live loop telemetry and the conversion model, trimming setpoints toward the lowest energy per tonne the plan allowed; purge and trip settings were read, never written.

  4. 04 evaporate.control SUCCEEDED continuous

    React-and-Evaporate · Held urea melt concentration at the granulation target from the online concentration analyzer and the evaporation model, moving evaporator setpoints inside the recipe window rather than waiting for the next lab assay.

  5. 05 granulate.control SUCCEEDED continuous

    Granulate-and-Prill · Tuned recycle ratio, spray and bed temperature from the live Granvis size distribution and moisture signal, each move validated on the twin first; the granule distribution stayed inside the spec band while recycle trended down.

  6. 06 condition.cool SUCCEEDED continuous

    Coat-and-Condition · Held dryer and cooler outlet temperature and moisture inside the storage window and set the anti-caking coating rate to the predicted caking risk for the product and the season; nothing moved outside the conditioning envelope.

  7. 07 inspect.granules SUCCEEDED continuous

    Defect-and-Inspect · 12 camera streams at the plant edge: granule size and shape, dust, caking precursors and broken prills on the granulator discharge, cooler and bagging belts; two dust spikes were attributed to one spray nozzle and Granulate-and-Prill corrected it inside its window.

  8. 08 approve.human APPROVAL 2 m 14 s

    Complex Orchestrator · The lowest-energy candidate wanted a purge-rate change on the ammonia loop beyond the bounded envelope. Held for the shift process engineer and the process-safety approver, who approved a smaller step with the safety-instrumented envelope untouched and a tighter review interval.

  9. 09 blend.bag SUCCEEDED continuous

    Blend-and-Bag · Held the NPK blend ratio to the nutrient spec from Granvis nutrient sensing, checked segregation on the belt and held bag weight to target; three overweight bags were attributed to one dosing gate and the gate was trimmed inside the approved window.

  10. 10 release.genealogy SUCCEEDED 1 m 10 s

    Quality-and-Conformance · Lot released: conversion and energy per tonne, melt concentration, granule distribution, hardness and moisture, nutrient assay, emissions readings, inspection events, bag weights and every human decision written to the lot genealogy and checked against spec and permit.

Compare

What is in each plan

Modules are licensed per unit on the Unit tier and bundled on Complex and Enterprise.

What is in each plan
CapabilityUnitComplexEnterprise
Units covered1All on siteAll in group
Agent modules1 of 11All 11All 11 + custom models
Nitrotwin twinNot includedIncludedCustom complex models
Plant-edge nodeIncludedPer unitFleet managed
Guarded DCS and APC writesScoped tagsAll scoped tagsGroup policy
Lot genealogyIncludedIncludedIncluded + export
Autonomy stagesShadow, assist, boundedShadow, assist, boundedShadow, assist, bounded
DeploymentPlant edge + cloudPlant edge + cloudOn-prem or private cloud
Outcome pricingOn requestAvailableStandard
Indicative price$22,000 / mo$140,000 / mo$900k – $12M 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.

tool-call stream · run_scn_01
  1. 05:10:02mes.read_order({ campaign: "U-4471", grade: "UREA-46", pack: "BAG-50" })
  2. 05:10:02ok grade spec locked · permit limits v4 · SIS envelope read-only
  3. 05:10:03historian.read_baseline({ unit: "AMM-1", window: "30d" })
  4. 05:10:03ok energy per tonne baseline loaded · catalyst age noted
  5. 05:10:04twin.simulate({ candidates: 24, horizon: "campaign" })
  6. 05:10:45ok best plan #9 · permit margin held · lowest energy per tonne
  7. 05:10:46policy.evaluate({ plan: 9, stage: "bounded" })
  8. 05:10:46ok 7 writes permitted · 1 write requires human approval
  9. 05:10:47dcs.write_setpoint({ tag: "AMM1.LOOP_T", basis: "conversion+energy" })
  10. 05:14:30ok conversion on target · energy per tonne trending down
  11. 05:20:00analyzer.read_concentration({ unit: "EVAP-2" })
  12. 05:20:00ok melt concentration inside window · evaporator setpoint held
  13. 05:31:12granvis.read_distribution({ unit: "GRAN-1", metric: "size_moisture" })
  14. 05:31:12ok distribution inside spec band · recycle ratio trimmed after twin check
  15. 05:31:13vision.stream({ cameras: 12, models: ["size_shape", "dust", "caking", "bag_defect"] })
  16. 06:02:48dust spike · cooler belt · attributed to spray nozzle 3
  17. 06:02:49defect.attribute({ event: "dust_spike", cause: "granulator.nozzle_3" })
  18. 06:02:49ok granulate notified · spray trimmed inside window · second spike, then clean
  19. 06:02:50dcs.write_setpoint({ tag: "COAT1.RATE", basis: "moisture+caking_risk" })
  20. 06:02:50ok coating rate set · moisture inside storage window
  21. 06:02:52approval.request({ action: "loop.purge", reason: "beyond bounded envelope" })
  22. 06:05:06ok approved by the shift process engineer and the process-safety approver · smaller step · SIS untouched
  23. 07:40:10plc.write_setpoint({ tag: "BLEND1.RATIO", basis: "nutrient+segregation" })
  24. 07:40:10ok nutrient ratio on spec · bag weight on target · gate 2 trimmed
  25. 09:18:30mes.release_lot({ campaign: "U-4471", genealogy: "full" })
  26. 09:19:40ok lot released · scenario complete

Reasoning trace

Goal, plan, thought, action, observation, exactly as the orchestrator would record it.

GoalRun campaign U-4471 from synthesis gas to bagged lot: ammonia …

Run campaign U-4471 from synthesis gas to bagged lot: ammonia loop at target conversion per unit of energy, melt at exact concentration, granules on spec with minimum dust and recycle, conditioned for storage, bagged to weight and released with full genealogy and emissions readings.

PlanSimulate the whole campaign on the twin first. Lead with conve…

Simulate the whole campaign on the twin first. Lead with conversion so evaporation gets the right melt and the granulator gets the right concentration. Verify emissions and the safety envelope continuously. Let the granulator chase the size band only inside the approved window; hold any ammonia-loop move beyond the envelope for a human.

ThoughtThe standing practice runs the loop and the granulator on fixe…

The standing practice runs the loop and the granulator on fixed setpoints and corrects after the assay. With live conversion, concentration and granule distribution the loop and the granulator can sit at target. The saving is energy, recycle and dust, not safety margin: the safety-instrumented envelope and the permit limits are never candidates for optimization.

Actiontwin.simulate(candidates=24) → plan #9: conversion on target, …

twin.simulate(candidates=24) → plan #9: conversion on target, melt concentration on spec, granule distribution inside the band, permit margin unchanged, lowest predicted energy per tonne.

ObservationPlan #9 wants a purge-rate change on the ammonia loop beyond t…

Plan #9 wants a purge-rate change on the ammonia loop beyond the bounded envelope. The policy engine routes that single move to the shift process engineer and the process-safety approver and permits the other seven writes.

ActionExecute plan #9 at the bounded stage: loop temperature, evapor…

Execute plan #9 at the bounded stage: loop temperature, evaporator, granulator, coating and blend writes proceed; emissions are verified continuously; the purge move waits.

ObservationApproval granted at 06:05:06 with a smaller step and the safet…

Approval granted at 06:05:06 with a smaller step and the safety-instrumented envelope untouched. Two dust spikes were attributed to spray nozzle 3 and corrected by Granulate-and-Prill inside the approved window.

OutcomeScenario complete. The lot is released with conversion and ene…

Scenario complete. The lot is released with conversion and energy per tonne, melt concentration, granule distribution, nutrient assay, emissions readings 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. Nitreum makes the stage explicit per loop, auditable and reversible at any time, and some decisions never leave the person. On the ammonia side every stage is safety-gated.

Autonomy stages and the human role at each
StageWhat the agent doesWhat the person doesWhen
ShadowPredicts conversion, concentration, granule quality, nutrient content, energy, emissions and off-spec risk; writes nothingRuns the plant as today; compares predictions to assay, historian and engineer outcomesFirst phase with every design partner
AssistRecommends bounded loop, evaporation, granulation, coating and blend moves with the evidenceA process engineer approves or rejects each oneAfter accuracy is proven against the baseline
BoundedWrites low-risk granulation, conditioning and blending setpoints inside documented recipe, quality, safety, emissions, equipment and rollback limitsApproves anything above the limits and every ammonia-loop move beyond the envelope; reviews the shift recordAfter the safety and ROI gates agreed up front
ExceptionsEscalates anything outside its envelope, safety-adjacent moves and release decisionsAlways decides these; the agent never doesPermanent
Accelerated computing

Physical AI, at the plant edge

A fertilizer complex is remote, continuous and often restricted from cloud links, so perception has to be local and fast. Nitreum runs GPU inference at the plant 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.

Jetson Orin / Thor · DeepStream · Metropolis · TensorRT · Holoscan

Edge perception

8 to 40 camera feeds per unit through DeepStream pipelines and TensorRT detectors and segmenters for granule size and shape, dust, caking, broken prills and bag defects; Holoscan fuses vision with analyzer, historian, emissions, acoustic and thermal streams at 1 to 10 Hz. Design target: 50 to 120 ms p95 inference.

Triton · NIM · NVIDIA AI Enterprise

Model serving

Vision, time-series, RAG and process-reasoning models behind one orchestrator, with canary deployment and per-site routing. NVIDIA AI Enterprise is the planned runtime for regulated deployments.

DGX / HGX · NeMo · Retriever · Guardrails

Training and reasoning (planned)

Granule vision, time-series prognostics and NeMo process-reasoning models fine-tuned from labeled images, process histories, engineer corrections and event logs. Design target: 2 to 5 model families and 500 to 2,000 GPU-hours per cohort.

Omniverse · Modulus · OVX

Digital twin (planned)

The Nitrotwin complex twin: physics-informed surrogates for the ammonia loop, evaporation, granulation, blending and emissions, with GPU-accelerated process and CFD sweeps run before a plant change.

Cosmos · Omniverse Replicator

Synthetic data (planned)

Rare dust, caking, wet-granule, broken-prill, lighting, belt-contamination, emissions-spike and upset scenarios generated and labeled as synthetic, validated against real plant events before a model learns from them.

cuOpt

Optimization

Throughput, recycle balancing, energy and emissions allocation and grade-changeover sequencing under constraints, recomputed on the GPU when the complex is disrupted.

RAPIDS

Data engineering

ETL over historian, quality, emissions and vision data for feature generation, quality correlation and cross-site benchmarks on the GPU.

Isaac · Jetson

Robotic handling

Bag palletizing, bulk loading, materials handling and inspection rovers, with perception and motion policies built and validated in simulation before they move a pallet.

Design targets

Product design targets

The numbers Nitreum is engineered against, from the GPU utilization plan.

50–120 ms p95 edge inference for granule size, shape, dust, caking and bag-defect frames
8–40 cameras per synthesis, granulation or bagging unit, 1080p or 4K at 15 to 30 FPS
1–10 Hz sensor fusion of vision, analyzer, historian, emissions, acoustic and thermal streams
2–5 model families fine-tuned per design-partner cohort
500–2,000 GPU-hours of fine-tuning per design-partner cohort
50k–250k synthetic rare-upset images and scenarios per cohort, labeled as synthetic

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 unit runs in shadow mode.

From the control room

Who this is built for

The three people in a fertilizer complex whose week changes when the process runs itself, and what each of them needs from Nitreum.

Economic buyer

Complex or Plant Operations Manager

Owns production rate, quality, yield and off-spec, energy, emissions and safety. Wants autonomous, higher-yield, lower-energy, lower-emissions production and a number that justifies the spend.

Champion

Process or Production Engineer

Owns synthesis, evaporation, granulation and blending recipes and process windows, energy and yield. Approves and corrects the agents in assist mode; those corrections train the models.

Approver

Sustainability, HSE and Energy Lead

Owns energy, CO2, NOx and ammonia-slip emissions, process safety, environmental compliance and the license to operate. Needs every action grounded, cited and written to an immutable record.

IT/OT and process safety approve the deployment; safety is a gating stakeholder for ammonia. 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

Nitreum reads and writes through the systems already in the control room. 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.

DCS, APC and historian

Reformer, ammonia loop, reactor and evaporator control and the plant historian

Setpoint reads, guarded writes, safety systems untouched

Granulation and prilling control

Granulator, prill tower, dryer, cooler and coating drum controls

Recycle, spray, temperature and coating setpoints inside windows

Online analyzers

Nutrient, moisture and concentration analyzers, NIR and granule analyzers

Continuous quality state, fused at the edge

Emissions sensing

CO2, NOx and ammonia-slip sensors and permit records

Read and verified; excursions predicted, limits never optimized against

Blending and bagging

NPK blender, dosing gates, bagging and loading lines

Ratio and weight setpoints inside approved windows

Cameras and robots

Fixed cameras on granule, belt and bagging stations; Isaac-driven palletizing and loading cells

Frames, defect events, bounded handling

Plant MES, LIMS and quality

Orders, grade specs, lab assays, lot genealogy

Targets in, release records out

Identity (planned)

SSO and role-based access; SOC 2 on the roadmap

Named approvers, scoped roles

See all integrations

Guardrails

An agent that can move a setpoint needs a leash

Nitreum writes to production equipment in a plant that handles ammonia. 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, through approved DCS or APC interfaces, inside per-tag rate and magnitude limits, at the autonomy stage the plant has set.
  • Safety systems are read, never written. Safety-instrumented systems, interlocks, trip settings and emissions permit limits stay with the plant's process-safety program and the people who own it.
  • Policy engine before every write. Autonomy stage, safety envelope, interlock state, grade, shift and operator presence are evaluated before a setpoint moves. Any ammonia-loop move beyond the envelope waits for a named approver.
  • Grounded and cited. Every recommendation carries a citation to the spec, procedure or permit 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 the plant's quality, emissions and process-safety record-keeping.
  • Hard fallback. If the edge node, the network or a model stops, the plant keeps running on its existing DCS, APC and safety systems.
  • Tenant and IP isolation. Recipes, process settings and plant IP never cross a customer boundary. On-prem and plant-edge deployment available.

Assurance posture

Assurance and certification status
ItemScopeStatus
SOC 2 Type I readinessCloud control plane QUEUED Roadmap, first 6 months
SOC 2 Type IICloud control plane QUEUED Roadmap, 6 to 12 months
Process-safety and emissions proceduresGrounding and record-keeping RUNNING Design basis for retrieval and the audit log
Per-tenant isolationData, memory, models RUNNING Design rule
On-prem and plant-edge deploymentIP- and safety-sensitive sites RUNNING Design option

Nitreum supports your process-safety, quality and emissions records. It is not a certification and it does not replace your safety program or management of change.

Read the security overview

Enterprise

Standardize autonomy across the group

One policy model, one audit trail and one set of yield, energy, emissions and granule-quality benchmarks across every complex in the group, with each site's recipes, process settings and emissions models kept private to that site.

Talk to us Enterprise details

  • Multi-site plant-edge fleet with shared model governance
  • Custom product, process and emissions 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 complex operations managers, process engineers and HSE leads 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, through the approved DCS or APC interfaces. Every design partner starts in shadow mode, where Nitreum predicts and writes nothing. Assist mode follows, where a process engineer approves each recommendation. Bounded autonomy on granulation, conditioning and blending loops comes after the safety and ROI gates agreed up front.

Get started

Pick your first product

Most complexes start with Synthex on the ammonia loop, Granulex on a granulator or Granvis for quality visibility. Tell us which number you need to move.