E2B raises $21M as 88% of Fortune 100 companies adopt its AI agent platform

E2B, a startup building cloud infrastructure specifically for artificial intelligence agents, has raised $21 million in Series A funding led by Insight Partners, a global software investor managing over $90 billion in assets. The round included participation from existing investors Decibel, Sunflower Capital, and Kaya, along with Scott Johnston, former CEO of Docker.

The funding comes as E2B reports that 88% of Fortune 100 companies have signed up to use its platform—a remarkable adoption rate that highlights how quickly large enterprises are embracing AI automation tools. The company has also added seven figures in new business just in the past month, according to CEO Vasek Mlejnsky, while processing hundreds of millions of computing sessions since October.

This rapid enterprise adoption reflects a fundamental shift in how companies are deploying AI. Unlike the chatbots and content generation tools that dominated early AI adoption, today’s enterprises are building AI agents—autonomous software programs that can execute complex, multi-step tasks including writing code, analyzing data, and browsing the web. However, these powerful capabilities create a significant infrastructure challenge that E2B has positioned itself to solve.

AI agents represent a fundamentally different computing paradigm than traditional software applications. While conventional programs follow predetermined paths, AI agents make autonomous decisions and often need to execute code they generate themselves. This creates serious security and scalability challenges that existing cloud infrastructure wasn’t designed to handle.

“Enterprises have enormous expectations for AI agents. However, we’re asking them to scale and perform on legacy infrastructure that wasn’t designed for autonomous agents,” Mlejnsky explained. “E2B solves this by equipping AI agents with safe, scalable, high-performance cloud infrastructure designed specifically for production-scale agent deployments.”

The core problem is that AI agents frequently need to run untrusted code—software they’ve written themselves that could potentially damage systems or access sensitive data. Traditional cloud platforms struggle to provide the isolation and security controls necessary to safely execute this code at enterprise scale. Industry data suggests fewer than 30% of AI agents successfully make it to production deployment, often due to these infrastructure limitations.

E2B addresses this challenge using Firecracker microVMs—lightweight virtual machines originally developed by Amazon Web Services—to create completely isolated environments for AI-generated code execution. Think of these as secure computing bubbles where AI agents can run potentially dangerous code without any risk to the broader enterprise system.

The company’s customer roster demonstrates the breadth of AI agent applications across industries. Search engine Perplexity uses E2B to power advanced data analysis features for Pro users, implementing the capability in just one week. AI chip company Groq relies on E2B for secure code execution in its Compound AI systems, while workflow automation platform Lindy integrated E2B to enable custom Python and JavaScript execution within user workflows.

The platform has also become critical infrastructure for AI research. Hugging Face, the leading AI model repository, uses E2B to safely execute code during reinforcement learning experiments. Meanwhile, UC Berkeley’s LMArena platform has launched over 230,000 E2B sandboxes to evaluate large language models’ web development capabilities.

These use cases illustrate why enterprises are willing to pay for specialized AI infrastructure. JPMorgan Chase, for example, has saved 360,000 hours annually through document processing agents, while industry leaders expect to automate 15% to 50% of manual tasks using AI agents. However, realizing these benefits requires infrastructure that can safely and reliably execute the code these agents generate.

E2B’s platform supports multiple programming languages including Python, JavaScript, and C++, and can spin up new computing environments in approximately 150 milliseconds—fast enough to maintain the real-time responsiveness users expect from AI applications. The platform can scale from 100 concurrent sandboxes on the free tier to 20,000 concurrent environments for enterprise customers, with each sandbox capable of running for up to 24 hours.

Enterprise customers particularly value E2B’s deployment flexibility. Companies can self-host the entire platform for free using the open-source version or deploy it within their own virtual private clouds to maintain data sovereignty—a critical requirement for Fortune 100 firms handling sensitive information. Advanced enterprise features include comprehensive logging and monitoring, network security controls, and secrets management capabilities essential for large-scale compliance requirements.

The company’s open-source approach has created additional competitive advantages. The E2B sandbox protocol has become a de facto standard, with hundreds of millions of compute instances demonstrating its real-world effectiveness. This network effect makes it difficult for competitors to displace E2B once enterprises have standardized on its platform.

E2B faces potential competition from cloud giants like Amazon, Google, and Microsoft, which could theoretically build similar functionality. However, the company has carved out a defensible position through its specialized focus on AI-specific use cases and open-source strategy.

“We don’t really care about the underlying virtualization technology,” Mlejnsky noted, explaining that E2B focuses on creating an open standard for how AI agents interact with computing resources. “We are actually partnering with a lot of these cloud providers too, because enterprise customers want to deploy E2B inside their AWS account.”

Alternative solutions exist but come with significant limitations. Docker containers, while technically possible for code isolation, lack the security characteristics required for production AI agent deployments. Building similar capabilities in-house typically requires 5-10 infrastructure engineers and at least $500,000 in annual costs, according to Mlejnsky—making E2B’s plug-and-play solution attractive even for large enterprises with substantial technical resources.

The funding comes at a pivotal moment for AI agent technology. Recent advances in large language models have made AI agents increasingly capable of handling complex, real-world tasks, while code generation assistants already produce at least 25% of the world’s software code. Microsoft’s recent workforce reductions while expecting AI agents to perform previously human-only work exemplifies this broader industry shift toward AI automation.

Insight Partners’ investment validates the market opportunity for AI infrastructure companies. “Insight Partners is excited to back E2B’s visionary team as they pioneer essential infrastructure for AI agents,” said Praveen Akkiraju, Managing Director at Insight Partners. “Such rapid growth and enterprise adoption can be difficult to achieve, and we believe that E2B’s open-source sandbox standard will become a cornerstone of secure and scalable AI adoption across the Fortune 100 and beyond.”

The company plans to use the funding to expand engineering and go-to-market teams in San Francisco, develop additional platform features, and support its growing customer base. Priority development areas include strengthening the open-source sandbox protocol as a universal standard and building enterprise-grade modules like secrets vault and monitoring tools.

E2B’s rapid adoption among Fortune 100 firms reveals a fundamental shift in how enterprises approach AI deployment. While much attention has focused on large language models and AI applications, the company’s success demonstrates that specialized infrastructure has become the critical bottleneck for AI transformation initiatives.

This trend reflects a broader maturation of enterprise AI. As AI agents transition from experimental tools to mission-critical systems, the underlying infrastructure requirements more closely resemble those of traditional enterprise software than consumer AI applications. Security, compliance, and scalability—not just model performance—now determine which AI initiatives succeed at scale.

For enterprise technology leaders, E2B’s emergence as essential infrastructure suggests that AI transformation strategies must account for more than just model selection and application development. The companies that successfully scale AI agents will be those that invest early in the specialized infrastructure layer that makes autonomous AI operation possible.

In an era where AI agents are poised to handle an ever-growing share of knowledge work, the platforms that keep those agents running safely may prove more valuable than the agents themselves.