AI Infrastructure Startup Modal Labs Raises $80 Million

Key Concepts

• Infrastructure layer for applications (generative media, large language models, etc.)
• Aggregation of underlying physical infrastructure (GPUs)
• Software layer for engineers to build on top (training, inference, batch processing, code execution)
• Capital efficiency through leveraging existing cloud infrastructure
• Developer experience as a key differentiator
• Demand from later-stage and enterprise companies
• Computational biotech and weather forecasting as emerging use cases

Platform Overview

The company builds the infrastructure layer for various applications, including generative media, large language models, and biotech platforms. This involves aggregating underlying physical infrastructure, specifically GPUs, from various sources globally. The core offering is a software layer that simplifies the process for engineers to build applications on top of this infrastructure. This includes support for training, inference, batch processing, and code execution.

Examples: Medha, lovable Sue, know ramp scale are mentioned as customers.

Differentiation from Existing Solutions

While there are similarities to services like AWS Lambda, the company differentiates itself by focusing on the specific needs of modern applications. Traditional infrastructure solutions like Lambda, Kubernetes, and Docker are not optimized for these workloads. The company has built a new stack from the ground up to address these challenges.

Relationship with Hyperscalers

The company partners with major hyperscalers like Azure. The hyperscalers provide the physical infrastructure and data centers, while the company focuses on innovating in the software layer above, specifically the developer experience. This allows engineers to build applications on top of the infrastructure more easily.

Funding and Growth Strategy

The company has raised over $80 million in funding, bringing the total to $111 million. The funds will be used to hire engineers to further develop the platform and invest in sales and marketing. The company is experiencing significant demand from customers, including later-stage and enterprise companies that are moving research prototypes into production.

Talent Acquisition

Finding the right talent, particularly engineers with expertise in infrastructure, is a challenge. Being located in New York provides some advantages, but the competition for skilled professionals remains intense.

Capital Efficiency

The company prioritizes capital efficiency by not owning the underlying physical infrastructure. Instead, it runs on top of existing cloud providers and neo clouds. This allows for faster expansion and reduces the capital burden.

Global Expansion and Use Cases

While most customers are currently in the US, the company is expanding globally. Lovable is mentioned as a customer that originated in Sweden. The company is seeing a wide range of use cases, including generative media, large language models, computational biotech (e.g., using models to cure cancer), and weather forecasting.

Conclusion

The company is building a specialized infrastructure platform for modern applications by aggregating GPUs and providing a developer-friendly software layer. By focusing on the software layer and partnering with existing cloud providers, the company achieves capital efficiency and rapid scalability. The company is experiencing strong demand and is expanding globally, with diverse use cases emerging across various industries.