- Austin-based Coder raised a $90M Series C led by KKR’s Next Generation Technology Growth Fund III, with participation from Qube Research & Technologies (QRT) and Uncork Capital.
- The company moves software development environments off local machines and into the cloud, giving enterprises governed, standardized workspaces for both human developers and AI coding agents.
- Coder reported 300% year-over-year bookings growth over the past four quarters and a 184% net dollar retention rate.
- The raise arrives as AI coding tools like Cursor and Claude Code push enterprise teams to rethink where and how code is written, stored, and secured.
Something structural is shifting in how enterprises build software, and a nine-year-old startup in Austin thinks it has the infrastructure layer that holds the whole thing together.
On April 1, 2026, Coder announced a $90 million Series C led by KKR, with Qube Research & Technologies (QRT), Uncork Capital, and existing investors also participating. The round is the company’s largest to date, arriving less than two years after a $35 million Series B2 in June 2024.
What Happened
KKR committed the lead investment through its Next Generation Technology Growth Fund III. The firm is not a passive backer — it is also a customer. According to the press release, KKR deployed Coder across more than 500 of its engineers as part of a push to standardize development environments and introduce AI tooling in a controlled way. Since that rollout, the firm has gone from zero AI-assisted code to a state where more than half of all commits occur within Coder-managed environments.
Ruchir Swarup, Chief Information Officer at KKR, put it directly: “Coder has fundamentally changed how we approach software development — particularly as we integrate AI into our workflows.”
QRT, the quantitative trading firm, took a similar path. It deployed Coder to more than 1,000 employees — roughly half its workforce — spanning engineers, data scientists, and analysts. Zohar Melamed, Head of Developer Experience at QRT, said the firm needed to “audit every LLM request, control access, maintain compliance” as it scaled agentic AI.
Ben Pederson, Managing Director at KKR, cited an internal view that more than 80% of enterprise developers already use or plan to use AI coding agents daily. “As enterprises scale the use of AI in development,” Pederson said, “they need infrastructure in a secure, standardized, and repeatable way.”
Why It Matters
The Series C is not just a fundraise — it is a signal about where enterprise software development is heading. The old model of every developer running their own local machine, with their own configuration, their own installed tools, and their own API keys sitting in dotfiles, is increasingly incompatible with how AI coding tools actually work.
AI agents are persistent. They run for hours. They write to multiple repositories simultaneously. They consume credentials and make network calls. Managing that on individual laptops is what Coder CEO Rob Whiteley called “a recipe for failure at best and breaches at worst.”
Coder’s answer is to move the development environment itself into the cloud — centralized, standardized, and governed — so that both humans and agents operate inside infrastructure the organization actually controls. The investment signals that large capital allocators now see that infrastructure layer as a durable enterprise bet, not a niche developer tool.
The company reported 300% year-over-year bookings growth across the past four quarters, 45% quarter-over-quarter growth in Q1, and a 184% net dollar retention rate — a metric that measures how much existing customers expand their spending over time. That last number is unusually high and suggests customers are not just renewing; they are adding seats and workloads.
Technical Details
Coder’s core product is an open-source platform that replaces local development setups with centralized, cloud-based workspaces. Enterprises provision these environments on their own cloud infrastructure or on-premises hardware, which means source code and credentials stay within the organization’s security perimeter.
The platform integrates with VS Code, JetBrains, Cursor, Windsurf, and other editors, so developers do not need to change their tooling — only where that tooling runs. Onboarding time drops because environments are templated and reproducible. Configuration drift across a large engineering team is eliminated.
Alongside the core product, Coder ships an open-source AI tool called Mux, which runs AI coding agents within development environments in isolated workspaces. Each agent gets its own sandboxed space, preventing the race conditions and file-state conflicts that arise when multiple agents work in a shared local environment.
The paid enterprise tier adds cybersecurity controls, MCP server restrictions, tool access management, agent activity audit logs, and development inference expense caps — governance features that compliance-heavy industries require before they will allow AI agents to touch production code.
Anthropic’s own engineers have used Coder to run Claude Code remotely in parallel multi-agent workflows, as detailed in a Coder blog post. The setup allows multiple agent runs to execute simultaneously across different tasks without local Docker overhead or Git state conflicts — a pattern that scales in ways local development cannot.
Who’s Affected
The most immediate audience is enterprise engineering teams at firms operating under strict compliance and security requirements — financial services, defense, pharmaceuticals, government, and telecom. These organizations have the most to lose from ungoverned AI agent activity and the most to gain from a centralized infrastructure layer that lets them adopt AI coding tools without surrendering control.
Individual developers working at those firms will encounter Coder as the environment their employer provisions rather than a tool they choose. The experience is designed to be transparent — editors and workflows remain unchanged — but the underlying compute and governance layer moves to the organization’s cloud.
The AI coding tools themselves — Cursor, Claude Code, GitHub Copilot, Windsurf — are also implicated. Coder does not compete with them; it provides the infrastructure those tools run on at enterprise scale. As Coder integrates with more AI coding surfaces, the boundary between IDE, agent runtime, and development infrastructure continues to collapse.
Smaller development teams and individual contributors are less directly affected for now. Coder offers open-source versions of its tools, but the enterprise governance features and the economics of the Series C are aimed squarely at large organizations with hundreds or thousands of engineers.
What’s Next
Coder has stated it will use the $90 million to advance platform features for enterprise AI workflows, deepen governance capabilities, and expand into Europe, Asia, and North America. The company’s growth trajectory — 148% year-over-year in Q1 alone — suggests it has significant capacity to deploy capital into both product and go-to-market expansion.
The broader developer infrastructure market is moving quickly. Remote development environments, once a convenience for distributed teams, are becoming a security and governance requirement in the AI coding era. Coder is one of the few companies with a production-grade open-source core, enterprise customers who are also lead investors, and a clear technical thesis for why local development fails at agentic AI scale.
For enterprise engineering leaders, the concrete next steps are narrow: evaluate whether your current development environment strategy accounts for AI agents running autonomously inside those environments. If the answer is no — if agents are currently operating on individual developer machines with ad hoc credentials and no audit trail — the Coder architecture represents one well-funded answer to that problem. The company’s open-source tools are available at coder.com, and the enterprise tier can be evaluated against the governance requirements your compliance and security teams will eventually impose regardless.
