Reactor control & DCS
Emerson DeltaVSiemens PCS 7Honeywell ExperionOPC UAModbus/TCP
Integrations
We meet your plant where it already is validated.
Polymorra reads and writes through the control system, instruments, and record systems you already qualified. No rip-and-replace, no parallel source of truth.
Integrations
Connects to the systems already running your plant
Polymorra reads and writes through the reactor DCS, PAT instruments, crystallizer and isolation skids, historians, and MES you already validated.
PAT & analytics
Mettler-Toledo FBRMPVM in-situ imagingReactIREndress+Hauser RamanNIRHPLC / XRPD imports
Isolation & drying
Nutsche filter-dryersCentrifugesAgitated dryersConical millsMicronizers
MES, historian & quality
Körber PAS-XEmerson SyncadeOSIsoft PILIMSeBR / QMS exports
Control layer
Your DCS stays in charge of safety
Polymorra never bypasses your interlocks. Setpoint moves are written through the same qualified path an operator would use, inside limits configured per train, and the DCS remains the ultimate authority.
- OPC UA and native DeltaV, PCS 7, and Experion connectors
- Read-only shadow mode available on day one
- Write authority granted per train, per parameter, per limit
- Instant handback and fail-safe on loss of the edge node
PAT layer
Instrument-native, vendor-neutral
Raman, FBRM, PVM, and NIR streams are ingested at native rate, time-aligned to the reactor telemetry, and stored with the batch — so a prediction can always be traced back to the exact spectrum and image it came from.
- Mettler-Toledo AutoChem FBRM, PVM, and ReactIR
- Endress+Hauser and Kaiser Raman
- NIR probes and at-line HPLC / XRPD imports
Record layer
MES, historian, and LIMS stay the system of record
Polymorra writes structured batch context and exception summaries back into PAS-X, Syncade, PI, and your LIMS, so quality review happens where your reviewers already work.
- eBR and QMS-friendly exception exports
- Historian tag mapping with unit-operation context
- LIMS result reconciliation against in-line predictions
API
A control API you can reason about
Everything the console does is available over the API: subscribe to batch state, read predictions with their evidence, and submit bounded control proposals for approval.
from polymorra import Plant
plant = Plant.connect("edge://site-04", verify_tls=True)
batch = plant.train("CR-2").active_batch()
state = batch.state() # fused PAT + telemetry estimate
print(state.supersaturation) # 1.28
print(state.d50_um) # 82.4
print(state.polymorph) # Form II (p=0.97)
for move in batch.proposed_moves():
print(move.parameter, move.target, move.rationale)
print(move.evidence) # spectra, images, telemetry refs
move.approve(signer="a.kaur")Edge runtime
Install at the train in an afternoon
The plant-edge runtime is a signed container image with a pinned model bundle. Install, point it at your OPC UA endpoint and PAT streams, and start in shadow mode.
$ polymorra edge install --site site-04 --train CR-2
✓ verified image signature (sha256:9f2c…)
✓ pinned model bundle crystallize-v4.2.1 (validated)
✓ connected opc.tcp://dcs.site-04.local:4840
✓ FBRM, PVM, Raman, NIR streams time-aligned
✓ audit trail initialised (21 CFR Part 11 mode)
→ autonomy level: SHADOW (no write authority)Plant-edge console
One screen for the batch that is running right now
Supersaturation, particle size distribution, predicted polymorph, impurity trajectory, and the next control move — with the evidence behind each number.
Polymorra plant-edge console showing live supersaturation, particle size distribution, predicted polymorphic form, and impurity trajectory for a running crystallization, with the recommended next control move.
Trust & compliance
Validated for GMP manufacturing
Polymorra is engineered for regulated drug-substance production: grounded outputs, immutable audit trails, graduated autonomy, and validation documentation from day one.
21 CFR Part 11
Immutable, validated audit log for every agent perception, recommendation, and control action, with e-signature-ready review flows.
GMP / ICH Q7
Change control, validated model versioning, and deployment documentation designed for drug-substance manufacturing.
SOC 2
SOC 2 Type I in progress, Type II on the roadmap; SSO/RBAC, encryption in transit and at rest. [ASPIRATIONAL]
Route & IP protection
Per-tenant isolation of routes, recipes, spectra, and crystal images. No cross-tenant training. On-prem option.
Graduated autonomy
Shadow → assist → bounded auto-control. Every autonomy level is explicitly configured, bounded, and revocable.
Grounded outputs
Every recommendation cites the spectra, images, telemetry, SOP, or specification it was derived from.
Plant-edge first
Inference runs on-site so control loops survive network loss; cloud is for training, fleet, and analytics.
Human-in-the-loop
Chemist and QA checkpoints are first-class: approve, correct, or reject — and every correction trains the models.
Enterprise
Built for multi-site pharma and CDMO networks
Multi-site deployment, plant-edge fleet management, custom route and crystallization models, validation support, and on-prem or hybrid installation with NVIDIA AI Enterprise.
Fleet of plant edges
Manage models, autonomy levels, and versions across every train in every site from one control plane.
Validation package
Qualification documentation, model version locks, and change-control artifacts your quality unit can review.
Outcome-linked pricing
Enterprise agreements can tie a component of fees to polymorph consistency, yield, or reprocessing reduction.
Dedicated engineering
Solutions engineers who speak crystallization, PAT, and GMP — not just software.
FAQ
Frequently asked questions
Both — in a deliberate order. Polymorra starts in shadow mode (perceive only), moves to assist (recommend, chemist approves), then to bounded auto-control on qualified trains where setpoint moves stay inside validated control windows. Every level is configured per train and revocable.
Polymorra is built for GMP (ICH Q7) and 21 CFR Part 11: immutable audit trail, validated model version locks with rollback, change control, grounded and citable outputs, and human-in-the-loop checkpoints. We ship a validation package and support IQ/OQ/PQ activities with your quality unit.
API or OPC access to your reactor DCS, PAT instruments (Raman, FBRM, PVM, NIR), crystallizer and isolation skids, and MES or historian — plus a train with enough campaign volume for clear ROI.
No. Per-tenant isolation of routes, recipes, spectra, and crystal images is enforced at the data and model layer. Cross-site benchmarks are opt-in and de-identified.
A typical wedge pilot instruments one train, runs shadow mode over a baseline campaign, then moves to assist mode with a defined polymorph, yield, or reprocessing success metric agreed up front.
Perception and control inference run at the plant edge. Cloud handles training, fleet management, and analytics — losing it degrades reporting, not the batch.
Send us your tag list. We will tell you what is possible.
Start with one crystallizer or reactor train. Prove polymorph, yield, and reprocessing ROI in a validation-friendly pilot. Then expand across the plant.