Intellegens is a machine learning software company, a spin-out from the University of Cambridge, with a unique data analysis technology that can speed up development times and reduce experimental workloads in the chemicals, life science, materials, and manufacturing sectors.
Easy Engineering: What are the main areas of activity of the company?
INTELLEGENS: Our expertise is in applying machine learning, a branch of artificial intelligence. We work on problems in scientific and engineering research and development.
Machine learning extracts value from data by using it to train a mathematical model so that the model captures key relationships within the data. The model can then be used predictively, for example, to compute likely outputs from a system for a new set of inputs, or to find the inputs that deliver a target output.
A problem with machine learning is that it usually fails when the training data used to generate the model is sparse (has lots of gaps) or noisy (contains inaccuracies). This makes it unsuitable for many real-world experimental applications.
The Intellegens Alchemite™ technology can handle sparse, noisy data. Intellegens has developed a software product and expertise that enables science and engineering teams to apply this methodology to challenges such as the design and development of formulations, materials, and chemical or manufacturing processes.
E.E: What’s the news about new products?
INTELLEGENS: Alchemite™ is now available as a commercial software product, recently enhanced in the Alchemite™ 2022 Autumn Release. The product is being used by a growing list of prestigious research organizations who develop formulated products, materials, or related industrial manufacturing processes. Examples include: global food producers; leading manufacturers of plastics, paints and coatings; additive manufacturing innovators; specialty chemicals companies; pharmaceutical manufacturers; producers of steels and other alloys; and aerospace organizations. Published case studies and papers in 2022 include work with NASA, Johnson Matthey, and AstraZeneca.
E.E: What are the ranges of products?
INTELLEGENS: Alchemite™ Analytics is web browser-based software that makes it easy to upload data, train a machine learning model, and then gain insights into the model through a wide range of graphical analysis tools. Alchemite™ Analytics helps scientists and engineers to apply machine learning to optimize products and processes and guide their experimental work.
Alchemite™ Engine provides data scientists with advanced API access to the Alchemite™ machine learning method, enabling them to integrate it with their existing machine learning and data analysis tools and workflows.
Alchemite™ Success is the Intellegens scientific service offering, which enables organizations to work with Intellegens experts to gain maximum value from machine learning technology. Usually, Intellegens scientists collaborate with customers on an initial pilot project, working out how to best apply the technology to the organization’s data and applications. As part of this process, customers are trained to become self-sufficient in using the software on future projects.
E.E: At what stage is the market where you are currently active?
INTELLEGENS: The application of machine learning to practical problems in the chemicals, materials, and manufacturing sectors is in its relatively early stages, owing to the challenges of handling real-world, sparse, noisy data and of deploying new methods effectively. But most major chemicals now companies have initiatives to investigate and apply machine learning in their R&D and its use is expanding rapidly with the advent of a tool that can overcome these barriers.
E.E: What can you tell us about market trends?
INTELLEGENS: Three major issues face all the businesses that we work with:
Costs, particularly of energy – the global crisis driving volatility in energy and raw materials costs means that these organizations need to urgently find changes in their products and processes that lower energy consumption and use less expensive ingredients, for which there are more reliable supply chains.
Sustainability – most major research organizations are setting themselves ambitious targets to contribute to global net zero emissions targets. This is leading them to look hard at the design of formulated products, chemicals, and materials, since carbon emissions associated with their production and processing are substantial. One construction chemical, concrete, alone contributes 4-8% of global emissions.
Speed of innovation – to meet these two challenges, and to win in highly competitive markets, there is an urgent need to reduce time-to-market for new products and for product and process improvements. Any technology that can provide extra insight and reduce the amount of time-consuming experimentation is of high value. That’s where Alchemite™ machine learning fits in.
E.E: What are the most innovative products marketed?
INTELLEGENS: The Alchemite™ method is highly innovative, the result of years of research at the University of Cambridge and of further development at Intellegens. But just as important is the software environment within which the method is delivered and its ability to make the method easy to apply in real-world industrial environments. The software is continually improved through close collaboration with customers. For example, in December 2022, our Alchemite™ Focus Group met, bringing together twenty user organizations as part of an on-going process to better understand and meet industry requirements.
E.E: What estimations do you have for 2023?
INTELLEGENS: At Intellegens, we now have a mature software product, delivering an innovative machine learning technology that can make a real impact on key business issues for chemicals, materials, and manufacturing businesses. We also have an established group of customers, many of whom are expanding their use of the software. We look forward to 2023 as the year in which Alchemite™ machine learning becomes a mainstream R&D technology.