2022-Korea-Showcase-Xinterra

PATRIC TEYSSONNEYRE: Hi, everyone. Thank you, Joanne, for the kind introduction. So, as Joanne introduced me, my name is Patrick Teyssonneyre, I'm one of the co-founders of Xinterra and the CEO of the company. And my connection to MIT is that I did the Sloan Fellows MBA program-- I'm part of the class of '19. And two of my co-founders are from MIT, as well, Professor Buonassisi, from MIT, is also a co-founder of Xinterra.

At Xinterra, we are on a mission to fundamentally change the development of sustainable materials to meet the urgent challenge of the climate crisis. The current materials R&D process is very slow, expensive, and limited. And also the materials that we use in our day to day, they work, they do their job. However, they are suboptimal, in terms of performance, cost, and sustainability.

And to solve this problem, we created our platform named XDF, or Xinterra Design Factor, that uniquely combines artificial intelligence systems, high throughput experimentation tools that we develop in-house, and deep material science expertise to help companies radically accelerate the development and application of new materials.

In this case here, a company spent seven months trying to develop a thermal insulating coating. They spent-- so they tried six to nine formulations, none of them worked. So they hired us to help solve this problem.

They shared their data with us, anonymized. We train our machine learning. Then in less than one month, we were able to help them to find four formulations that worked very well-- actually much better than their expectations. So basically, this happened because they were initially exploring if you see the data circled by red, they are exploring that space.

And the machine learning basically told us, look, you are exploring the wrong space. You should start exploring the space circled in yellow. So then the algorithms predicted, and recommended, specific formulations that are behind each one of these dots. The customer reproduced these formulations in the lab, make the trials, and then we were able to find these four formulations. All of this in less than one month.

In this case, a company hired us-- it's a company, they produce a specialty chemical. They hired us-- they wented us to recombine the 15 ingredients that they use in this chemical, in a way that preserves exactly the same performance. Doesn't need to improve, but couldn't get worse, as well.

So in a way that this would reduce the cost of this product by 5%. It's a very commoditized industry-- 5% is a lot of money for this industry. And just to have a reference, 10% is something-- and it would be something unimaginable for them.

So the same thing-- we received this data, we train our machine learning, and in less than one month, we were able to find two formulations that reduced the cost in more than 20%, and keeping exactly the same performance.

And in this case here, we removed 15-- sorry, three of the 15 ingredients. Two were the most expensive. A third one was-- this ingredient was really bad for the environment-- in terms of sustainability. Then, the remainder 12, we recombined them to reach this result.

So much more lower cost, this formulation, performs the same way, and is much more sustainable.

We are also using our platform-- so in this case, the combination of machine learning and high throughput exploitation tools, to develop a portfolio of materials for carbon capture. This is a representation of one of the rigs-- it's an environmental chamber that we developed, to measure the CO2 absorption performance of many different materials that we are screening.

We already-- so they start us going to [INAUDIBLE] to measure one sample per day, maximum two. We already-- these tools that we developed are measuring already 15 per day, so we have a 50 times increase in the output just in three months. So we are creating a massive database of carbon capture materials that we use this database to train our artificial intelligence.

But very soon, we start recommending new materials, new formulations or combinations of existing materials, that could be developed-- and that would be much more efficient in terms of carbon capture-- efficient, cost-efficient, as well, on absorbing CO2. And one of the ideas is to integrate these materials into paints and coatings, textiles, as well. Because part of our vision is how to empower every human being walking the surface of our planet to become a carbon capture agent.

So these are some examples of the data that we are generating. So you can see the curves of the dozens of different materials that we are generating, and understanding all the impacts of each one of the parameters, how they're correlated, what causes [INAUDIBLE], we well, and we are starting to build some models to predict the behavior of new materials, as well.

In terms of the business model, we have two business models-- the first one, we provide service to companies. This service could be project specific, like the two first cases that I presented. And for companies that want to access our platform on a more regular basis, and that once we have more autonomy, we can build tailored AI models to them.

We have a good track record here. So we already have several customers from startups to multinational companies with more than $20 billion US in sales.

And then, the other business model is, as I mentioned, so we are using our platform to develop a portfolio of materials IP, to license to the companies in the near future. And we are starting by materials for carbon capture.

So why Korea? In terms of-- so it's clearly a country that recognizes the urgency to combat climate change. Therefore, the companies have urgency, as well, to launch sustainable materials. And companies are also eager to digitize, even further.

In terms of culture-- so we love the pali pali, so we are a pali pali of materials. So fast paced, we like this fast pace. That's why we are developing materials using this platform. So cultural alignment-- it's a culture that is technology and knowledge centric.

And because we are based in Singapore, same time zone, easy to access. So these are some of the reasons.

We are looking for collaborations in the service model. We're looking for new customers in these industries-- chemicals, polymers, paints and coatings, pulp and paper, and personal care. And for the CO2 capture initiative, we are looking for early stage partners to co-develop scaleup manufacture and commercialize the materials that we are designing.

And these are the priority industries-- oil and gas, chemicals, textiles, paints and coatings, and appliance.

Thank you for our time. If you have any questions, we'll have in the other room, in our booth. Thank you.