Applied Computing raises $20M for AI model to streamline oil & gas plant operations

Applied Computing, founded in 2023 and based in London, has secured $20 million（約32億円） in Series A funding to accelerate deployment of its AI model for the oil, gas, and petrochemical industry. KBR, an engineering giant, led the round, with Databricks Ventures also participating. The startup targets a critical operational challenge: energy facilities typically have thousands of sensors measuring temperature, pressure, velocity, and viscosity, yet operators use less than 8% of the available data when making operating decisions, according to CEO Callum Adamson.

The startup's core product is Orbital, a foundation model designed to solve the data integration problem. Unlike conventional large language models that predict the next word, Orbital combines three components: a time series model to analyze sensor readings over time, a physics-based model to respect the laws of thermodynamics and chemistry, and a language model to interpret facility documentation and operator activity. The system allows technicians to simulate how changes in one part of a facility affect the rest of its operations. Adamson told TechCrunch that the fundamental challenge is "getting those three data sources to talk to each other in real time."

Orbitals promises significant speed gains. Applied Computing claims the model can flag anomalies, investigate their causes, and model whether a proposed fix could create problems elsewhere in the facility—all within minutes. Adamson states that Orbital compresses investigations that previously took days or weeks into seconds, helping operators reduce energy consumption and maintain output. The startup has already achieved traction: it has grown from stealth to double-digit millions in annual recurring revenue in under 18 months. Adamson confirmed that Orbital is in use at some large, publicly listed upstream oil and gas, downstream refining, and petrochemicals companies, though he did not disclose the customer count.

Partnerships validate the technology. Indian energy company Wipro is a partner, and KBR has integrated Orbital into its INSITE 3.0 digital platform for energy projects and is deploying it for ammonia production. Applied Computing is also working with a major U.S. upstream operator and plans to announce a partnership with a European oil major in the coming weeks. The competitive landscape includes established players like AspenTech (which sells simulation and AI-powered modeling software), AVEVA (physics-based process simulation and optimization), Cognite and Seeq (data-layer tools). Adamson argues that Applied Computing's moat is neither data access nor energy-domain expertise but rather the ability to attract tier-one AI researchers to build a model that outperforms competitors. He noted that real operational data from refineries and petrochemical plants is generally unavailable publicly, giving Applied Computing an advantage through deployments with actual facilities.

The company will use the $20 million（約32億円） to expand internationally, hire researchers and engineers, and deepen customer deployments. Applied Computing has opened an office in Houston, joining its London headquarters and operational hub in Bengaluru, to serve two existing North American customers and prepare for further expansion. Adamson indicated that expansion into the Middle East is also underway. The company's geographic and hiring strategy reflects confidence that Orbital can scale beyond its initial North American and European deployments to become a global solution for energy operations.

Applied Computing enters the energy sector at a moment when industrial AI is becoming more sophisticated but also more competitive. The core problem Adamson identifies—that facilities collect extensive sensor data but use less than 8% of it for operational decisions—reflects a longstanding challenge in industrial automation: integrating diverse data streams and knowledge domains faster than human operators can manage. The startup's insight is that this is fundamentally an AI engineering problem rather than a data or domain-expertise problem. By assembling top AI researchers and combining multiple specialized model types (time series, physics-based, and language models) rather than relying on a single large language model approach, Applied Computing positions itself as solving a technical barrier that competitors like AspenTech, AVEVA, Cognite, and Seeq have not fully cracked.

The KBR partnership is strategically significant beyond the funding itself. KBR's integration of Orbital into its INSITE 3.0 platform and its use of the model for ammonia production suggests not just validation but distribution and customer access. Adamson's emphasis on operational data—data from real working plants that cannot be replicated in simulation—points to a key competitive moat. The combination of double-digit millions in annual recurring revenue in under 18 months and partnerships with established industrial players and energy majors indicates that the market is willing to adopt the technology. However, Applied Computing must now navigate entrenched vendors and prove that its AI engineering advantage can scale across different plant types and geographies as it expands to the Middle East and deepens North American presence.