Several years ago, Fabrizio Del Maffeo and a core team from Imec, a Belgium-based nanotechnology lab, teamed up with Evangelos Eleftheriou and a group of researchers from IBM Zurich Lab to develop a computer chip. Unlike conventional chips, theirs was destined for edge devices, especially those with AI workloads, because Del Maffeo and the rest of the team saw that most offline, advanced computing hardware was inefficient and expensive.
After incubating a startup – Axelera AI – to commercialize their chip technology within the blockchain company Bitfury Group, Del Maffeo and team raised capital from VCs, including Imec’s venture arm, Imex.xpand. Innovation Industries led a $27 million Series A in Axelera AI that closed this week with the participation of imec.xpand and the Federal Holding and Investment Company of Belgium. In addition, the Netherlands Enterprise Agency has awarded Axelera AI a loan of $6.7 million on behalf of the Ministry of Economic Affairs and Climate.
“One of the main challenges we face is the availability and accessibility of AI to different groups and industries. Some companies would benefit from AI but lack the knowledge or technical expertise to integrate its benefits into day-to-day operations,” Del Maffeo told londonbusinessblog.com in an email interview. “We are developing the AI platform to help overcome this barrier to entry… [by] delivering a breakthrough, easy-to-use and scalable technology with superior performance and efficiency at a fraction of the cost of existing players to accelerate computer vision and natural language processing at the edge.”
Axelera is developing AI acceleration maps and systems for use cases such as security, retail and robotics that it plans to sell through partners in the business-to-business edge computing and Internet of Things sectors. The cards and systems include Axelera’s Thetis Core chip, which uses in-memory computing for AI calculations – “in-memory” refers to performing calculations in RAM to reduce latency caused by storage devices.
Axelera is also making software to manage its chip, which Del Maffeo claims will be “fully integrated” with leading AI frameworks — such as PyTorch and TensorFlow — when made available.
Axelera’s test chip for accelerating AI and machine learning workloads. Image Credits: Axelera
“We are democratizing AI access,” said Del Maffeo. “When our product is launched… we strive to [deliver] a chip that contains the power of a full AI server.”
However, Axelera still has a long way to go before it hits the market. The company didn’t produce its first-ever test bench chip until last December, and Alexea doesn’t expect to be able to deliver to customers until sometime in the first half of 2023.
Nor is it the first company to pursue an in-memory architecture for edge devices. NeuroBlade, which raised $83 million in capital last October, is developing chips that combine both computing power and memory into a single data processing hardware block. MemVerge, GigaSpaces, Hazelcast, and H20.ai also provide in-memory solutions for AI, data analytics, and machine learning applications.
But despite the fact that Axelera is both pre-market and pre-revenue and takes into account the progression of venture debt rounds, Del Maffeo believes the company is well-positioned to gain a foothold in the custom AI market. -potato chips. He notes that Jonathan Ballon, the former VP and general manager of Intel’s edge AI and Internet of Things group, joins Axelera as chairman. And Del Maffeo points out that Axelera continues to rent aggressively, with nearly 85 employees in Europe, remotely and in the company’s offices in Eindhoven and Milan and R&D centers in Leuven and Zurich.
Following an expansion in the US and Taiwan in the coming months, Del Maffeo expects Axelera to enter in early 2023 with 130 to 140 employees.
“While the pandemic led to several shortages in the chip industry, we have been fortunate to see significant growth since we launched in 2021,” said Del Maffeo. “While we are not disclosing our burn rate, we can share that we are well positioned to raise another investment round in 2023, significantly larger than our Series A, and we are already getting interest, including from more US investors, to join us. help take the business to the next stage… We are carefully building our customer base and partner ecosystem with a curated cohort of companies that have already shown strong interest in our AI platform.Later this year, we will also open a unique collaboration opportunity for leading companies to get early become adopters of our AI platform.”
Assuming Axelera can deliver on its promises, it could make some serious money. The edge AI hardware market is: projected to grow from 920 million units in 2021 to 2.08 billion units in 2026 – a lucrative increase. according to an estimationthe AI chip market will be worth $73.49 billion by 2025 alone.