TQ Ventures led the round, with participation from FPV Ventures, Redpoint, and Unusual Ventures. The investment values Railway as one of the hottest infrastructure startups to emerge during the AI boom, tapping into developers’ frustration with the complexity and cost of traditional platforms like Amazon Web Services and Google Cloud.
“As AI models get better at writing code, more and more people are asking the age-old question: Where and how do I run my applications?” Jake Cooper, the 28-year-old founder and CEO of the rail company, said in an exclusive interview with VentureBeat. “The last generation of cloud primitives was slow and outdated, and now with AI moving everything faster, teams simply can’t keep up.”
The funding represents a significant acceleration for a company that has charted an unconventional path through the cloud computing industry. Railway had raised just $24 million in total before this round, including a $20 million Series A from Redpoint in 2022. The company now processes more than 10 million deployments a month and handles more than a trillion requests through its terminal network – metrics that rival much larger, better-funded competitors.
Why three-minute posting times are becoming unacceptable in the age of AI coding assistants
Railway’s pitch is based on a simple observation: The tools developers use to deploy and manage software were designed for a slower era. A standard build and deployment cycle using Terraform, our industry-standard infrastructure tool, takes two to three minutes. This delay, which was acceptable, has become a critical bottleneck as AI coding assistants like Claude, ChatGPT, and Cursor can generate working code in seconds.
“When divine intelligence is available and can solve any problem in three seconds, integrating systems becomes bottlenecks,” Cooper told VentureBeat. “What was really great for humans to deploy in 10 seconds or less is now servings on the table for customers.”
The company claims that its platform delivers deployments in less than one second – fast enough to keep up with AI-generated code. Customers report a ten-fold increase in developer speed and up to 65 percent cost savings compared to traditional cloud providers.
These numbers come directly from enterprise customers, not internal benchmarks. Daniel Lobaton, chief technology officer at G2X, a platform serving 100,000 federal contractors, measured deployment speed improvements seven times faster and cost reductions 87 percent after moving to rail. His infrastructure bill dropped from $15,000 a month to about $1,000.
“Work that would take me a week on our previous infrastructure, I can do on the rail in about a day,” Lobaton said. “If I wanted to build a new service and test different architectures, it would take a long time in our old setup. On the railways, I can run six services in two minutes.”
Inside the controversial decision to ditch Google Cloud and build data centers from scratch
What sets Railway apart from competitors like Render and Fly.io is the depth of its vertical integration. In 2024, the company made the unusual decision to abandon Google Cloud altogether and build its own data centers, a move that echoes Alan Kay’s famous quote: “People who are really serious about software should build their own hardware.”
“We wanted to design the devices in a way that we could build a different experience,” Cooper said. “Having complete control over the network, compute and storage layers allows us to build and deploy rings very quickly, the kind that allows us to move at ‘agent speed’ while maintaining 100 percent smooth ride in the city.”
This approach paid off during recent widespread outages affecting major cloud providers – and Railway remained online throughout.
This comprehensive control enables pricing that undercuts hyperscalers by approximately 50 percent and newer cloud startups by three to four times. Rail fees per second for actual compute usage: $0.00000386 per gigabyte per second of memory, $0.00000772 per vCPU per second, and $0.00000006 per gigabyte per second of storage. There are no fees for idle virtual machines – a stark contrast to the traditional cloud model where customers pay for available capacity whether they use it or not.
“The conventional wisdom is that larger companies have economies of scale to offer better prices,” Cooper noted. “But when they’re charging virtual machines that would normally be idle in the cloud, and we’ve designed everything specifically for much greater density on those machines, you have a huge opportunity.”
How 30 employees built a platform that generates tens of millions in annual revenue
The railroad company achieved its massive size with a team of just 30 employees, generating tens of millions in annual revenue – a revenue-per-employee ratio that would be extraordinary even for established software companies. The company’s revenue has grown 3.5 times in the past year and continues to expand by 15 percent month-on-month.
Cooper stressed that the fundraising was strategic and not necessary. “We are living in default, and there is no reason for us to raise money,” he said. “We grew up because we see a huge opportunity for acceleration, not because we need to survive.”
The company hired its first sales representative just last year, and employs only two solutions architects. Nearly all of Rail’s 2 million users discovered the platform through word of mouth, with developers telling other developers about a tool that already works.
“We basically did the standard engineering thing: If you build it, they will come,” Cooper recalls. “And to some extent, they came.”
From Side Projects to Fortune 500 Deployments: The Unexpected Expansion of a Railroad Company
Despite its grassroots development community, Railway has had significant inroads into larger enterprises. The company claims that 31 percent of Fortune 500 companies now use its platform, although deployments range from company-wide infrastructure to individual team projects.
Notable clients include loyalty program company Bilt; Intuit subsidiary GoCo; TripAdvisor cruise critic; and MGM Resorts. Kernel, a Y Combinator-backed startup that provides AI infrastructure to over 1,000 companies, runs its entire customer-facing system on rail for $444 per month.
“At my previous company Clever, which was sold for $500 million, I had six full-time engineers running just AWS,” said Rafael Garcia, chief technology officer at Kernel. “Now I have six engineers, all focused on the product. Rail is exactly the tool I wish I had in 2012.”
For enterprise customers, Railway offers security certifications including SOC 2 Type 2 compliance and HIPAA readiness, with Business Associate Agreements available upon request. The platform provides single sign-on authentication, comprehensive audit logs, and the option to deploy within the customer’s existing cloud environment through a “bring your own cloud” configuration.
Enterprise pricing starts at custom tiers, with specific add-ons for extended log retention ($200 per month), HIPAA BAAs ($1,000), enterprise support with SLOs ($2,000), and dedicated VMs ($10,000).
The startup’s bold strategy to compete with Amazon, Google and a new generation of cloud competitors
Railroad is entering a crowded market that includes not only large-scale cloud providers – Amazon Web Services, Microsoft Azure, Google Cloud Platform – but also a growing array of developer-focused platforms like Vercel, Render, Fly.io, and Heroku.
Cooper sees rail competitors divided into two camps, neither of which has fully committed to the new infrastructure model that AI requires.
“The hyperscalers have two competing systems, and they haven’t fully bought into the new model because their old revenue stream is still printing money,” he noted. “They have this huge amount of money coming in from people who are providing virtual machines, and they’re using maybe 10 percent of it, and they’re still paying for everything. How interested are they really in moving forward with a new experience if they don’t really need to?”
Compared to startup competitors, Railway stands out by covering the entire infrastructure package. “We’re not just containers, we have primitives for virtual machines, stateful storage, virtual private networks, and automated load balancing,” Cooper said. “And we wrap all this in a ridiculously easy-to-use UI, with agent primitives so agents can move 1,000 times faster.”
The platform supports databases including PostgreSQL, MySQL, MongoDB, and Redis; Provides up to 256TB of persistent storage with over 100,000 I/O operations per second; It allows deployment in four global regions, including the United States, Europe, and Southeast Asia. Enterprise customers can scale to 112 vCPUs and 2TB of RAM per service.
Why are investors betting that artificial intelligence will produce 1,000 times more software than exists today?
The railroad’s fundraising reflects broader investor enthusiasm for companies able to benefit from the AI programming revolution. As tools like GitHub Copilot, Cursor, and Cloud become standard tools in developer workflows, the amount of code being written – and the infrastructure needed to run it – expands dramatically.
“The amount of software that will be brought online over the next five years is unfathomable compared to what was there before. We’re talking about a thousand times more software,” Cooper predicted. “This all has to happen somewhere.”
The company has already integrated directly with AI systems, building what Cooper calls “loops through which Cloud can connect, invoke deployments, and analyze infrastructure automatically.” Railway released the Modular Context Protocol Server in August 2025 that allows AI coding agents to deploy applications and manage infrastructure directly from code editors.
“The developer’s idea is dissolving before our eyes,” Cooper said. “You no longer have to be an engineer to be able to engineer things, you just need to think critically and be able to analyze things with a systems ability.”
What does a railroad company plan to do with $100 million and no marketing experience?
The railroad plans to use the new capital to scale its global data center, grow its team beyond 30 employees, and build what Cooper described as a go-to-market operation for the first time in the company’s five-year history.
“One of my mentors said you raise money when you can turn the business around,” Cooper explained. “We have built all the foundations required to scale indefinitely; what’s holding us back is just talking about it. 2026 is the year we play on the global stage.”
The company’s investor list reads like a who’s who of infrastructure developers. Angel investors include Tom Preston-Werner, co-founder of GitHub; Guillermo Rauch, CEO of Vercell; Spencer Kimball, CEO of Cocrush Labs; Olivier Baumel, CEO of Datadog; and Gauri Lalu, co-founder of Linear.
The timing of the railway expansion coincides with what many in Silicon Valley see as a fundamental shift in how software is made. Coding assistants are no longer just an experimental curiosity, they have become essential tools that millions of developers rely on every day. Every line of code generated by AI needs somewhere to run, and incumbent companies, says Cooper, are too committed to their existing business models to take full advantage of the present moment.
Whether Railway can translate developer enthusiasm into sustained institutional adoption remains an open question. The cloud infrastructure market is filled with promising startups that have failed to break the stranglehold of Amazon, Microsoft, and Google. But Cooper, who previously worked as a software engineer at Wolfram Alpha, Bloomberg, and Uber before founding Railway in 2020, seems unfazed by the scale of his ambition.
“In five years, the railways [will be] “The place where software is created and developed,” he said. “Deploy instantly, scale infinitely, with zero friction. This is the prize worth playing for, and there’s no greater prize on offer.”
For a company that built a $100 million business by doing the opposite of conventional startup wisdom-no marketing, no sales team, no hype-the real test starts now. The railway company spent five years proving that developers would find a better mousetrap on their own. The next five will determine whether the rest of the world is ready to join us.
