Evan Lee Salim
The significant investment underscores a growing recognition within heavy industries of the transformative potential of artificial intelligence to unlock efficiencies, enhance safety, and drive better operational decision-making in complex environments. Founded in 2023, Applied Computing has rapidly emerged from stealth mode to address a critical pain point in the energy sector: the vast underutilization of operational data within industrial facilities. Modern oil rigs, refineries, and petrochemical plants are veritable data factories, housing thousands of sensors that continuously monitor parameters ranging from temperature and pressure to velocity and viscosity. Despite this abundance of information, facilities routinely make critical operating decisions based on less than 8% of the data available to them, according to Callum Adamson, co-founder and CEO of Applied Computing. This staggering statistic highlights a profound challenge in a sector where optimal performance is paramount for economic viability, environmental stewardship, and safety.
The Data Conundrum in Industrial Operations
The oil, gas, and petrochemical industries are characterized by intricate processes and highly capital-intensive assets. A single facility can span vast areas and involve countless interconnected systems, each generating real-time data streams. The problem is not a lack of data collection, but rather the immense difficulty in integrating, contextualizing, and analyzing this heterogeneous information quickly and effectively. Operators face a triple threat of data fragmentation: combining raw sensor readings, deciphering complex engineering documentation, and applying sophisticated physics and chemistry models. These disparate data sources often reside in siloed systems, use varying formats, and lack standardized ontologies, making real-time synthesis a formidable task.
Traditional analytical tools and human-led processes struggle to keep pace with the velocity and volume of this industrial data. Investigations into anomalies, performance deviations, or potential equipment failures can stretch for days or even weeks, leading to costly downtime, suboptimal resource allocation, and increased risk. The global digital oilfield market, valued at approximately $23.7 billion in 2022 and projected to grow significantly in the coming years, is a testament to the industry’s desperate need for advanced solutions to this data-tracking problem. However, the fragmentation across legacy systems, proprietary software, and diverse operational technologies has historically presented a significant hurdle to comprehensive data utilization. As Adamson succinctly put it to TechCrunch, "It’s getting those three data sources to talk to each other in real time. That’s the real key." This integration challenge is precisely what Applied Computing aims to solve with its innovative AI model.
Orbital: A New Paradigm for Industrial AI
At the heart of Applied Computing’s offering is Orbital, a proprietary foundation AI model designed specifically for the nuanced demands of the energy and petrochemical sectors. Unlike large language models (LLMs) that predict the next word in a sequence, Orbital is engineered to predict the state of a facility by combining three distinct yet interconnected AI components: a time series model, a physics-based model, and a language model. This multi-modal approach allows Orbital to process and understand the complex interplay of various factors influencing an industrial operation.
The time series model analyzes historical and real-time sensor readings, identifying patterns and anomalies over time. The physics-based model integrates fundamental engineering principles and chemical reactions, ensuring that predictions adhere to the immutable laws governing physical processes within the plant. This is a critical differentiator, as it prevents the AI from generating physically impossible or implausible outcomes, a common challenge with purely data-driven models in industrial contexts. Finally, the language model processes engineering documentation, operational manuals, and even technician notes, providing crucial contextual understanding of equipment specifications, operating procedures, and historical incident reports. By synthesizing these three data sources, Orbital can not only analyze sensor readings but also factor in equipment constraints and operator activity, offering a holistic view of the facility’s health and performance.
Beyond merely understanding the current state, Orbital empowers technicians to run sophisticated simulations. This "what-if" modeling capability allows operators to test the potential repercussions of a change in one part of a facility on the rest of its operations, all within a matter of minutes. This proactive approach to problem-solving and optimization marks a significant leap forward from traditional methods, which often rely on slow, manual calculations or limited simulation tools.
The Accelerated Path to Operational Excellence
Applied Computing’s core value proposition revolves around speed and precision. The company asserts that Orbital can drastically compress the time required for critical operational tasks. For instance, an investigation into a flagged anomaly that previously might have taken days or even weeks of manual data correlation and expert analysis can now be completed in seconds. This rapid diagnostic capability is crucial for minimizing downtime, preventing catastrophic failures, and optimizing resource consumption. By quickly identifying root causes and modeling effective solutions, operators can reduce energy use, maintain consistent output levels, and enhance overall operational safety. In an industry facing increasing pressure to improve efficiency and reduce its environmental footprint, these capabilities translate directly into tangible economic and sustainability benefits.
The market has responded enthusiastically to this promise. Applied Computing has demonstrated remarkable traction since its inception, moving from stealth mode to generating "double-digit millions" in annual recurring revenue in under 18 months. This rapid commercial success is a strong indicator of the acute need for such advanced AI solutions within the energy sector. CEO Callum Adamson confirmed that Orbital is already deployed and in use at several "large, publicly listed" companies across upstream oil and gas, downstream refining, and petrochemical operations, though specific customer names remain confidential.
Strategic Partnerships and Global Expansion
The Series A funding round and the participation of KBR and Databricks Ventures are strategic validations of Applied Computing’s technology and market approach. KBR, a global leader in engineering, procurement, and construction for the energy and government sectors, brings deep industry expertise, an extensive client network, and a direct channel for integration. Indeed, KBR has already incorporated Orbital into its INSITE 3.0 digital platform for energy projects and is actively utilizing the product for ammonia production, a critical process in the chemical industry. This partnership provides Applied Computing with invaluable access to operational data, industry best practices, and introductions to a broader base of potential customers, significantly accelerating its market penetration.
Databricks Ventures’ involvement further highlights the strategic importance of data intelligence platforms in enabling cutting-edge AI applications. Databricks’ expertise in data warehousing, data lakes, and AI/ML platforms aligns well with Applied Computing’s mission to leverage complex industrial data for predictive insights.
Beyond KBR, Applied Computing has forged alliances with other significant players, including Indian energy company Wipro. Adamson also revealed ongoing collaborations with a "major U.S. upstream operator" and hinted at an upcoming partnership announcement with a prominent European oil major in the coming weeks. These partnerships are crucial for demonstrating the versatility and scalability of Orbital across diverse operational contexts and geographical regions.
Navigating a Competitive Landscape
Applied Computing operates in a market that, while ripe for innovation, is not devoid of established players and emerging competitors. Entrenched industrial software suppliers like AspenTech and AVEVA have long offered simulation and AI-powered modeling solutions for upstream, refining, and chemical operations. AspenTech, for example, is known for its Aspen HYSYS, a process simulation software, while AVEVA provides physics-based process simulation, optimization, and "what-if" modeling for industrial plants. Furthermore, more focused AI startups such as Cognite and Seeq target the data layer, assisting facilities in analyzing industrial data and applying AI to design workflows.
Despite this competitive environment, Adamson asserts that Applied Computing’s primary moat is not merely access to industrial data or process knowledge, but rather its unique ability to assemble world-class AI researchers to build a foundation model like Orbital. He argues that the challenge is fundamentally "an AI problem. It’s not a data problem, and it’s not an energy problem." This perspective suggests that while domain expertise is vital, the core innovation lies in the advanced AI methodologies and talent required to develop such a sophisticated, multi-modal model. Adamson provocatively suggests that "If you’re a tier-one AI researcher, where are you going to work? I don’t think Shell’s on that list," implying that specialized AI startups are better positioned to attract and retain top AI talent compared to traditional energy giants.
Another critical differentiator lies in the quality and nature of the operational data Orbital ingests through its deployments. Adamson emphasized that real operational data from refineries and other energy facilities is generally not publicly available, and simulated data, while useful, cannot fully reproduce the complexities and nuances of a live, working plant. The direct access to proprietary, real-world operational data through client deployments and partnerships like KBR provides Applied Computing with an invaluable feedback loop, continuously enhancing Orbital’s accuracy and robustness.
Future Growth and Global Ambitions
The $20 million Series A funding will serve as a catalyst for Applied Computing’s ambitious growth plans. The company intends to leverage the capital to fuel international expansion, significantly expand its research and engineering teams, and pursue new deployments with a broader spectrum of energy clients.
In a tangible step towards its global aspirations, Applied Computing recently announced the opening of a new office in Houston, Texas, a strategic move that places the company closer to its existing customers in North America, a pivotal market for the energy sector. This new base complements its London headquarters and its operational hub in Bengaluru, India, creating a global footprint that supports its diverse client base and talent pool. Looking ahead, the company also has plans for expansion into the Middle East, another critical region for oil and gas production, underscoring its commitment to becoming a global leader in industrial AI.
Applied Computing’s journey represents a compelling narrative of how advanced AI, when meticulously tailored to the specific challenges of heavy industries, can unlock unprecedented levels of efficiency, safety, and strategic insight. As the energy sector navigates complex transitions towards sustainability while striving for operational excellence, foundation models like Orbital are poised to play an increasingly central role in shaping its future.
