CurieOS Platform by Orbital Industries: status and timeline

CurieOS Platform

ML platform for materials discovery that automates design and experimentation. Licenses to firms accelerating R&D in catalysts, pharma and batteries.

PilotStage 3 of 5

Licensing to BASF for automotive catalyst research (2026); expanding to pharmaceuticals and other industries.

Updated Aug 2026Checked 10 Oct0 updates this week

Milestones

Next · BASF catalyst research deployment

CurieOS platform developed2024Complete.
BASF licensing agreement signedAug 2026Complete.
BASF catalyst research deploymentTarget 2026Current milestone.
Pharmaceutical industry partnershipsTarget 2027Not yet reached.

Most important, last 3 months

BASF Environmental Catalyst and Metal Solutions deploys Orbital's CurieOS platform for catalyst researchAug 2026

Upcoming

2026BASF catalyst research deployment (next)
2027Pharmaceutical industry partnerships

Current obstacles

Domain-specific customisationDifferent chemical fields (catalysts vs pharmaceuticals) have different design constraints; the platform must adapt without losing generalisation.
Regulatory and safety validationNew materials discovered by AI must be validated for safety, manufacturability and compliance; the business model depends on partner acceptance.

Physics limits

Model quality depends on training dataAI predictions are only as good as the chemistry data available for training; novel regions of chemistry space are harder to predict.

How it works

4 parts

Model training

Learning from data

CurieOS is trained on historical chemistry data, experimental results and simulations to predict molecular properties and behaviour.

Design generation

AI-proposed candidates

The platform generates novel molecular structures predicted to have desired properties, far exceeding what human chemists alone would explore.

Experiment planning

Smart prioritisation

Autoresearch agents select which candidates to synthesise and test physically, choosing based on predicted impact and experimental uncertainty.

Iterative refinement

Closed-loop learning

Physical test results feed back into the model, refining predictions and guiding the next round of design and experimentation.

Spec sheet

Spec	CurieOS Platform
Maximum atoms per simulation	100,000+^R (reported)
R&D acceleration factor	10-100x vs traditional methods^E (estimated)

^R reported by the company · ^E our estimate

Update log

1 update

Aug 2026

Major: BASF Environmental Catalyst and Metal Solutions deploys Orbital's CurieOS platform for catalyst researchPressCommercial

About Orbital Industries

The team behind CurieOS Platform

Orbital Industries

AI materials discovery, GPU cooling, molecular simulation

Orbital Industries uses AI to discover novel materials and chemicals. It developed PFAS-free cooling fluids for GPU data centres and licenses CurieOS for catalyst research. Founded 2022 by former DeepMind researchers.

Orbital's Orb simulator and autoresearch agents design materials in months, not years.
Discovered PFAS-free GPU cooling fluid in months (vs 10 years, $100M traditionally).
Series B ($50M, May 2026) funds data-centre scale-up and CurieOS licensing to BASF.

orbitalindustries.com (opens orbitalindustries.com)Company profile →

Founded: 20224 yrs
Headquarters: United Kingdom
Status: Private
Valuation: private
Staff: ~50May 2026
Raised: $66M2 rounds
Last round: Series B · $50MMay 2026
Works in: MaterialsAI materials discoveryComputing
Coverage: 2 programs · 3 updateslatest 1 Aug 2026checked 10 Oct
Lead investors: PluralRadical Ventures
Partners: BASF Environmental Catalyst and Metal Solutions (opens basf.com)
Investors: Nvidia (opens nvventures.nvidia.com)Plural (opens plural.ai)
People: Jonathan Godwin (opens linkedin.com)Chief Executive OfficerJames Gin-PollockChief Technology OfficerDaniel MiodovnikChief Operating Officer

Spec

CurieOS Platform

Maximum atoms per simulation

100,000+^R (reported)

R&D acceleration factor

10-100x vs traditional methods^E (estimated)