FlexAI, a Paris-based startup, has exited stealth mode with a $30 million (€28.5 million) seed round led by Alpha Intelligence Capital (AIC), Elaia Partners, and Heartcore Capital. The company aims to revolutionize AI compute infrastructure, making it more accessible and efficient for developers.
The company's founders, Brijesh Tripathi and Dali Kilani, are industry veterans with impressive resumes, including stints at NVIDIA, Apple, Tesla, and Intel. The startup aims to build a layer of software intelligence, abstraction, and orchestration into the AI compute stack, enabling developers to seamlessly utilize multiple hardware configurations without complex code changes.
The current landscape of AI compute is complex and cumbersome, requiring developers to have extensive knowledge of infrastructure before they can even begin. FlexAI aims to change this by creating a cloud service that provides developers with "virtual heterogeneous compute." This approach abstracts away the underlying hardware, allowing developers to focus on building and training AI models without worrying about the intricacies of GPUs and chip architectures.
FlexAI's universal AI compute platform will enable developers to utilize multiple hardware configurations without the hassle of manual code changes. By partnering with hardware leaders like AMD, Intel, and NVIDIA, as well as cloud providers such as AWS and Google Cloud, FlexAI will offer developers a seamless experience.
The company addresses three major obstacles hindering AI innovation: the compute supply bottleneck, the global skills shortage, and the complex, unreliable processes for building and deploying AI models. "We believe in AI's transformative power to solve some of humanity's biggest problems, but it will require a 1000x magnitude more compute to be able to realize this vision," said Brijesh Tripathi, CEO and co-founder of FlexAI. "The availability of AI compute today is limited to a select few. Our vision is to unlock access to compute for the many."
The company's first commercial product, launching later this year, will be an on-demand cloud service, providing developers with flexible access to AI compute resources.
FlexAI's goal is to bring AI compute infrastructure to the same level of simplicity as general-purpose cloud computing. The company's concept of "universal AI compute" allows developers to focus on building, training, and using models while FlexAI manages the underlying infrastructure, failures, recovery, and reliability.
The startup plans to act as an aggregator of demand, renting hardware through traditional means and securing preferential prices from Intel and AMD to spread across its customer base. This approach sets FlexAI apart from competitors like CoreWeave and Lambda Labs, which focus primarily on NVIDIA hardware.