In a strategic move, Meta has bolstered its AI team with expertise from former engineers at Graphcore, enhancing its capabilities in cutting-edge technology.
In Brief
A specialized group of engineers, previously with Graphcore, has joined Meta to lend their expertise in AI networking chips.
In a recent report, Reuters highlighted the departure of ten individuals from Graphcore during late 2022 and early 2023, who then began their journey with Meta in the subsequent months.
These new recruits will play a pivotal role in advancing Meta’s AI and machine learning initiatives within the company’s data centers.
The company has successfully onboarded a talented group of engineers based in Oslo, who previously contributed to AI networking solutions at Graphcore, as reported. Reuters report .
According to findings by Reuters, ten professionals shifted from Graphcore to Meta after Graphcore ceased operations at its Oslo location due to restructuring, as documented on their LinkedIn profiles.
Jon Carvill, representing Meta, verified these new additions, explaining that the company has recently integrated a cadre of highly skilled engineers from Oslo into its infrastructure division. Their specialized knowledge in supercomputing systems is set to enhance Meta's AI and machine learning operations across its data centers.
The new engineers transitioned to Meta following the closure of Graphcore’s Oslo office, a decision made during a restructuring process last October amid growing competition from major players like Nvidia and AMD.
Graphcore achieved a significant valuation milestone in 2018, becoming a unicorn valued at $1.7 billion after securing $200 million in its Series D funding round, with investors like BMW and Microsoft backing it. In 2020, despite raising an additional $222 million and reaching a valuation of $2.77 billion, Graphcore faced challenges when Microsoft opted to utilize Nvidia’s GPUs instead of their own for cloud infrastructure development in Azure.
Currently, Meta is channeling substantial resources toward AI technology to keep pace with rival tech companies. There’s a tangible internal demand for AI infrastructure as various teams across the organization seek to leverage this technology. Meta has also been utilizing AI for targeted advertising and moderating prohibited content on its platforms.
While Carvill did not disclose specific projects, insiders have indicated to Reuters that Meta is developing an in-house division dedicated to creating diverse chip designs aimed at enhancing AI performance, including specialized network chips. These chips facilitate data transfer across networks and are produced by leading companies like Nvidia and Intel.
In addition, Meta is in the process of developing a sophisticated computing chip intended for training AI models, enabling them to make decisions and generate responses, although this technology is not anticipated to be ready until 2025.
Read more:
Disclaimer
In line with the Trust Project guidelines Please remember that the details presented here are not meant to serve as legal, financial, investment, or any form of professional advice. It's essential to only engage with investments that you are comfortable with and to consult a financial expert if you have any uncertainties. For additional guidance, we recommend reviewing the terms and conditions as well as support resources offered by the issuer or advertiser. At MetaversePost, we strive for accurate and unbiased journalism, but be mindful that market fluctuations can occur unexpectedly.
