London’s Applied Computing raises £15m for energy AI

TL;DR:

• Applied Computing, a London company building a foundation model for energy operations, has raised about £15 million (€17.4 million / $20 million) led by engineering group KBR, with Databricks Ventures joining as a new investor.
• The round funds US expansion through a new Houston office, alongside deeper investment in AI research.
• The company has appointed former Shell AI leader Dan Jeavons as president and says a first partnership with a European oil major will be announced within weeks.

Another London AI company has found its niche defensible enough to attract strategic money. Applied Computing, founded in 2023 by Callum Adamson and Dr Sam Tukra, has raised about £15 million (€17.4 million / $20 million) led by KBR — the US engineering group it already builds products with — alongside new investor Databricks Ventures. The round follows a seed round of roughly £9 million (€10.7 million) in May 2025.

The company’s pitch is specialisation. Its Orbital platform combines language models with time-series, physics and chemical engineering models, built for upstream, downstream and petrochemical operations rather than adapted from general-purpose AI. The company claims Orbital can use all available data from downstream energy facilities, against 8% captured by traditional methods. KBR has already shipped a product on it: INSITE 3.0, launched in March, runs on Orbital.

The investor mix tells the strategy. KBR brings decades of process engineering data and global distribution; a multi-year agreement makes the products exclusive to the sector. The Houston office puts the company in the US energy industry’s home city, while operations run from Bengaluru and headquarters stay in London. Ex-Shell AI leader Dan Jeavons joins as president — a hire aimed squarely at operator credibility.

The vertical foundation model bet

Applied Computing is a test of a thesis with wider UK relevance: that domain-specific foundation models — grounded in physics and sector data rather than web text — can win against general-purpose systems in industries where errors are expensive. For the UK’s AI ecosystem, it is a reminder that credible frontier plays need not be consumer chatbots; London-built industrial AI with strategic corporate backing is a viable route.

Looking forward

The funding backs commercial deployment with global energy operators and expanded AI research under Dr Tukra. The promised European oil major partnership, expected within weeks, will be the next signal of whether the vertical model translates into flagship customers.