Applied Computing Lands $20M as KBR Doubles Down on Its Energy AI Bet

Applied Computing has raised $20 million in a Series A round led by KBR, the engineering giant that made its first-ever AI investment in the London startup four months ago. Databricks Ventures joined as a new backer. The money funds a new Houston office and faster deployment of Orbital, Applied Computing’s foundation model for energy operations, across refineries, petrochemical plants and LNG terminals worldwide.

The round lands months after KBR’s original bet looked unusual on paper: a 36,000 person engineering contractor taking a board seat inside a three year old AI startup. This time KBR is doubling that wager, and a second outside investor is putting fresh money behind the same conviction.

KBR and Databricks Back a Second Bet in Four Months

The $20 million round, about €17.4 million, follows a £9 million seed raise in May 2025 led by Stride.VC and Repeat.vc, worth roughly $10.7 million at the time. Add the two rounds together and Applied Computing has now pulled in close to $30 million since 2023. Its valuation was not disclosed in either round.

Callum Adamson, the company’s chief executive and co-founder, framed the new capital as continuity rather than a fresh start.

It’s our mission to provide operators with a foundation model that unlocks advantage at scale while delivering pathways to production that are safer, more efficient and far less carbon intensive.

Greg Conlon, KBR’s chief digital and development officer, said the firm wanted to keep scaling what it had already started. “We’re very excited about what this technology could unlock across the full lifecycle for multiple industries, and we’re thrilled to make this investment in Applied Computing to spur future technologies,” he said.

Julien Debard, Databricks’ director of energy and utilities, said his firm’s decision to join came from watching the product work, not from a pitch deck. “We’ve seen the calibre of the technology and the strength of the team behind it, and how quickly operators can now move from experimentation to real operational value on the platform,” Debard said.

A Board Seat Turns Into a Bigger Check

The relationship between the two companies did not start with this round. KBR’s own announcement in March called its investment its first strategic investment in an AI company, paired with a board seat and a multi year joint development deal to build exclusive AI products for energy.

The sequence of events shows how fast that first bet turned into something bigger.

1. May 2025: Applied Computing closes a £9 million seed round led by Stride.VC and Repeat.vc.
2. March 17, 2026: KBR launches INSITE 3.0, described in its own release as a physics-informed AI adviser for plant operators, built on Orbital.
3. March 23, 2026: KBR announces a strategic investment in Applied Computing and takes a board seat.
4. July 2026: KBR leads a $20 million Series A, joined for the first time by Databricks Ventures.

Dan Jeavons, a former Shell AI executive who now serves as Applied Computing’s president, called the INSITE 3.0 launch a turning point at the time. “This is a true game-changer for the industry,” he said in KBR’s announcement.

Why Do Refineries Run on Just 8% of Their Data?

Most refineries and petrochemical plants already collect far more sensor data than they use. Applied Computing’s pitch is that operators make decisions with less than 8% of the information sitting in their own systems, because sensor readings, engineering documents and physics rarely get combined fast enough to matter in the moment.

Founded in 2023, Applied Computing targets facilities where thousands of sensors track temperature, pressure, velocity and viscosity at once. Orbital combines a time series model, a physics based model and a language model to read that mess and predict the state of a plant, rather than simply predicting the next word the way a general purpose chatbot would.

Jeavons described the reasoning behind that design in an earlier interview explaining Orbital’s architecture, arguing that operators should not have to choose between explainable models and accurate ones. “Surely, we can bring these two modeling approaches together, and that’s exactly what we’re trying to do with Orbital,” he said.

• 8% of available plant data gets used in a typical operating decision, according to Adamson.
• 90% outperformance over previously benchmarked software, by Applied Computing’s own figures.
• 18 months from stealth to double digit millions in annual recurring revenue, the company says.

Adamson has described the core engineering challenge in blunter terms. “It’s getting those three data sources to talk to each other in real time. That’s the real key,” he told TechCrunch.

Houston Joins London and Bengaluru on the Map

Applied Computing now runs three locations: London as headquarters, an operations hub in Bengaluru, and the new Houston office. Adamson said the Texas base puts the company closer to two existing customers in North America, with an expansion into the Middle East also in the works.

The Houston opening follows an earlier expansion into India, where Applied Computing set up its Bengaluru hub for industrial AI to support a growing customer base there. Together the three offices give the company a foothold across the three regions where most of the world’s refining and petrochemical capacity sits.

The company said the new funding will go toward a specific set of goals rather than general overhead:

• Opening the commercial office in Houston, its third location after London and Bengaluru.
• Expanding research and engineering headcount to build out Orbital’s model suite.
• Deepening deployments with energy customers already running Orbital in production.
• Pursuing planned expansion into the Middle East alongside its North American customer base.

Adamson also said Applied Computing is working with a major U.S. upstream operator and expects to announce a partnership with a European oil major within weeks, on top of existing ties to Wipro and KBR.

AspenTech and AVEVA Still Hold the Floor

Applied Computing is not moving into empty territory. TechCrunch noted the company is entering a market with entrenched industrial software suppliers as well as more focused AI startups chasing adjacent problems.

Where PhysicsX and Cognite focus on design and monitoring, Applied Computing is aiming squarely at live operational decision making inside running plants, a narrower but harder problem to get wrong.

What Applied Computing Still Won’t Confirm

Not every number in Applied Computing’s pitch has been checked by outside parties, and the company has kept several figures close.

• What we know: The $20 million round closed with KBR leading and Databricks Ventures joining, annual recurring revenue reached double digit millions inside 18 months, and Orbital already runs inside KBR’s INSITE 3.0 platform and an ammonia production deployment.

• What’s unconfirmed: Applied Computing’s valuation, its exact customer count, which Adamson has declined to share, and a separate claim that running Orbital across 55 refineries worldwide could cut emissions by more than the United Kingdom’s total share, which has not been independently verified.

Applied Computing has published its own figures on data usage on its account of the industry’s untapped operational data, arguing that most of what energy facilities generate never gets used at all. That gap, real or overstated, is the entire basis for KBR’s second check.

Adamson said the European oil major deal should be announced within weeks, adding a third continent to a customer map that already spans North America and Asia.