Cerebrium Raises $8.5 Million Seed Round to Scale Serverless AI Infrastructure Platform
Cerebrium, a serverless AI infrastructure startup founded by Michael Louis and Jonathan Irwin, has raised US$8.5 million in a seed funding round led by Gradient Ventures, with participation from Y Combinator, Authentic Ventures, and several strategic angel investors.
The company, originally founded in Cape Town in 2021 and now based in New York City, provides a high-performance, serverless platform for building, deploying, and scaling multimodal AI applications. These applications include workloads such as voice agents, video generation models, fine-tuning of large language models, and large-scale batch inference. Cerebrium’s goal is to eliminate the infrastructure complexity typically associated with these AI workloads.
Cerebrium offers serverless compute infrastructure, including both CPU and GPU support, that automatically scales based on usage. The platform is optimized for low-latency performance, including minimized cold start times, and supports features such as multi-region deployments, advanced observability tools, and built-in compliance capabilities. Customers are charged based on actual compute consumption, removing the need to manage or provision servers manually.
Among its early adopters are companies like Tavus, Deepgram, and Vapi, all of which use Cerebrium to support high-volume AI-powered operations. These companies rely on Cerebrium to ensure stability, speed, and scalability without having to build out and maintain dedicated infrastructure internally.
The seed funding will be used to expand the engineering team, develop new platform features, and continue scaling the company’s infrastructure to meet growing enterprise demand. Cerebrium plans to deepen its support for real-time inference, improve performance for multimodal input processing—including text, audio, and video—and add enhanced tools for security, monitoring, and deployment.
The idea for Cerebrium was born out of the founders’ personal frustrations with the fragmented ecosystem of cloud tools for AI developers. They experienced long development cycles, unreliable performance, and high infrastructure costs when building AI applications using traditional cloud providers. Cerebrium aims to solve these issues by offering a streamlined, unified platform that helps developers move from prototype to production faster and more efficiently.
Currently operating with a lean team of four full-time engineers, Cerebrium is already generating millions in annual recurring revenue from its growing base of customers. The company’s minimalist approach to team structure and its developer-first mindset have resonated with startups and AI teams looking for infrastructure that is both powerful and easy to use.
With this latest round of funding, Cerebrium is set to accelerate product development and broaden its presence in the enterprise AI market. The company is also focused on expanding globally, improving its ability to serve customers in different regions with low latency and high availability.
As AI models grow larger and more complex, the demand for robust, scalable infrastructure continues to rise. Cerebrium positions itself as a forward-thinking platform designed to meet the performance, cost-efficiency, and compliance needs of the next generation of AI applications. By offering a fully serverless environment optimized specifically for AI, the company is reducing friction for developers and enabling teams to focus more on building great products rather than managing backend systems.
