About Axelera AI
- Founders: Fabrizio del Maffeo (Co-Founder and CEO)
- Founded in: 2021
- Employees: 75
- Money raised: 10 million
- Ultimate goal: To spread AI around the globe and make the world a greener, safer place.
Artificial intelligence (AI) is the future, but to use this software in our day-to-day lives there is a need for faster, cheaper, and more efficient hardware. Start-up Axelera AI has developed a test chip that could make this possible: the Thetis Core. A fast memory computer chip that could help implement AI in our lives like never before. The start-up is just a little bit more than a year old, but has grown rapidly and is already planning on announcing its first version of the chip in the fourth quarter of 2022. The young enterprise also won a Gerard & Anton Award. In this episode of start-up-of-the-day, co-founder and CEO Fabrizio del Maffeo talks about his innovative product.
What is the Thetis Chip?
“To understand Thetis, I would like to start with the AI problem. Artificial intelligence and in particular neural networks are very different from other algorithms. If you look inside this network, you will see that a big part of the operations is vector-matrix multiplications: you multiply a vector, which is a column of numbers, with a table of numbers. This represents 70 to 90 percent of all calculations the hardware has to do. The rest are more sophisticated calculations. If you want to speed up your network, you have to speed up this part, so you automatically speed up everything else. Our technology, which is an in-memory computer chip, can speed up this process to make it more efficient. Our chip immediately makes all the calculations in one cycle. Typically a computer can do 16 or 32 calculations per cycle. Our test chip can do 260,000 calculations in one cycle. That’s because we do the calculation inside the memory. We modify the memory just to do this simple calculation of multiplication and accumulation which speeds up the process.”
Who stands to benefit from this chip?
“Our chip delivers high performance levels and enhanced usability at a fraction of the cost of what is on the market. In the first place, we want to create a technology that is accessible to anyone who designs hardware products. They need high levels of performance in order to scale up their applications. That could be medical devices or devices for retail outlets. We are targeting business to business. Then I would say practically anyone after that. We want to enable a thousand start-ups and established companies to scale up the performance of AI with a powerful tool that allows them to adopt AI in a wide variety of applications. For example, you could incorporate real intelligence into drones to monitor crops or even fires. Or you can go shopping and pay without having to stop at the checkout. The possibilities are endless.”
When did you start the company?
“We started on July 9th, 2021. We got our investments and from that moment we started to scale very fast. The idea started already in 2017. I was an engineer for almost 20 years in computing companies, covering different roles from development to management. I started looking at this trend of bringing intelligence from the cloud to the edge. To give more and more intelligence to devices. I saw there was a growing demand and interest in a solution to accelerate this process, so I started to see an opportunity there. I started talking with people and two years later the company was founded. Now, we are going to launch the product announcement before the end of the year. And we want to deliver the first version of the product to selected customers in the first half of next year.”
What problems did you have to overcome to make this chip?
“We are bringing to market a new technology and each day you discover something new. We are also doing this with a completely new team. So, you have to build new technology and build a new team. A virtual team of people to keep together in a moment when we had no money. I needed a hundred million before reaching a breakeven point and in Europe, this kind of project is considered too ambitious. A lot of investors told me I was in the wrong place, to take a flight and go to the United States and try it there. We didn’t give up and eventually, we found investors. I’m still here. We have 75 people now. Things are improved.”
How hard was it to find financial support?
“It was extremely difficult and it’s still extremely difficult to convince people to give you millions if you don’t have a product. I don’t want to play down the work of other start-ups, but if you develop software, which most start-ups do, you can actually have this proof of concept fairly readily.
We needed 10 million euros to start with just to prove our concept. We did have a top-level theoretical simulation that was expensive, but investors are invariably skeptical. We are also pitching for a market that is growing but it isn’t in place yet. Everyone knows that this market will be huge, but is isnt at the moment. So investors feel they should wait. And you are competing with some big companies, which is another problem.”
What makes your product better than what’s already on the market?
“It’s the unique performance per area and power consumption. You can have high performance, at the fraction of the cost of power consumption. AI is consuming a lot of power so it’s good to have a technology, which is in principle greener compared to others. It’s also cheaper than other chips. That way you can democratize AI and make it accessible for everyone.”
What’s the ultimate goal?
“For us, it’s to spread the use of artificial intelligence to make the world a safer and greener place. We firmly believe that AI can be used for good causes and that if we can spread the use of AI, we can improve the quality of life for everyone. You can save time because you can focus on things that create more added value. And to achieve this goal, we have got to democratize AI. In the end, we have given our technology to millions of companies and engineers so they can use their creativity to unleash the power of our chip to help create these new solutions. AI could really change the whole world.”
