Railway secures $100 million to challenge AWS with AI-native cloud infrastructure
Key Takeaways
- Railway, a San Francisco cloud platform, raised $100 million in a Series B round led by TQ Ventures, with FPV Ventures, Redpoint and Unusual Ventures also participating, according to VentureBeat.
- The company says it now serves two million developers, processes more than 10 million deployments monthly, and handles over one trillion requests through its edge network, per VentureBeat.
- Railway built its own data centers after leaving Google Cloud in 2024, and it prices compute by the second rather than charging for idle capacity, undercutting major cloud providers by roughly 50 percent, VentureBeat reports.
A Fast-Growing Challenger Emerges From Word of Mouth
Railway’s $100 million Series B, reported by VentureBeat, marks a sharp acceleration for a company that had previously raised only $24 million in total, including a $20 million Series A from Redpoint in 2022. The new round positions Railway as one of the more closely watched infrastructure startups to surface during the current wave of AI-driven software development, according to VentureBeat.
What makes the trajectory notable is how Railway got here. VentureBeat reports the company has not spent money on marketing, instead growing to two million developers largely through word of mouth. Jake Cooper, Railway’s 28-year-old founder and chief executive, told VentureBeat that the company hired its first salesperson only last year and currently employs just two solutions engineers, alongside a total headcount of around 30 people generating tens of millions of dollars in annual revenue, per VentureBeat.
Cooper framed the raise as opportunistic rather than urgent, telling VentureBeat that Railway was “default alive” and did not need the capital to survive, but wanted to move faster on a market opportunity it sees opening up. VentureBeat reports the company grew revenue 3.5 times over the past year and continues to expand at 15 percent month over month.
Why Deployment Speed Is Becoming a Bottleneck in the AI Economy
The core argument Railway is making, as relayed by VentureBeat, is that the infrastructure tools developers rely on to ship software were built for a much slower pace of work. VentureBeat notes that a standard build-and-deploy cycle using Terraform, described as the industry-standard infrastructure tool, takes two to three minutes — a delay that becomes more conspicuous now that AI coding assistants such as Claude, ChatGPT and Cursor can generate functional code within seconds.
Railway claims its own platform can complete deployments in under one second, a figure VentureBeat frames as central to the company’s pitch to enterprises. Customer-reported outcomes cited by VentureBeat include a tenfold increase in developer velocity and cost savings of up to 65 percent relative to traditional cloud providers. One cited example, Daniel Lobaton, chief technology officer at G2X, told VentureBeat his infrastructure costs fell from $15,000 per month to approximately $1,000 after migrating, alongside deployment speeds he measured as seven times faster.
This matters beyond one company’s product roadmap. As AI tools compress the time needed to write code, the remaining friction in getting that code running in production becomes more visible — and more costly to businesses racing to ship AI features. VentureBeat’s reporting suggests this dynamic is pushing some technical teams to reconsider long-standing dependence on hyperscalers like Amazon Web Services and Google Cloud, at least for certain workloads. For readers tracking the broader AI economy, this is part of a pattern in which infrastructure spending is being scrutinized more closely as AI adoption scales, with efficiency and cost control becoming competitive differentiators rather than afterthoughts.
The knock-on effects for markets, including crypto-adjacent technology sectors, are indirect but worth noting. Blockchain and crypto infrastructure projects often rely on the same categories of cloud compute, storage and networking that Railway is targeting, and shifts in pricing power or vendor competition in cloud infrastructure can influence the operating costs of decentralized applications, node operators and Web3 startups that lease traditional cloud capacity. VentureBeat’s report does not address crypto specifically, and no such claims should be read into it, but the broader trend of infrastructure cost compression is relevant to any technology sector, crypto included, that depends on cloud spending as a major line item.
Vertical Integration and Enterprise Traction
According to VentureBeat, Railway’s move to build its own data centers in 2024, after exiting Google Cloud entirely, is what separates it from competitors such as Render and Fly.io. Cooper described this as giving the company full control over network, compute and storage layers, which VentureBeat reports allowed Railway to remain online during recent outages that affected major cloud providers.
That control also underpins Railway’s pricing model. VentureBeat lists Railway’s per-second charges as $0.00000386 per gigabyte-second of memory, $0.00000772 per vCPU-second, and $0.00000006 per gigabyte-second of storage, with no fees for idle virtual machines — a structure the company says undercuts hyperscaler pricing by about 50 percent and newer cloud startups by three to four times.
VentureBeat reports that 31 percent of Fortune 500 companies now use Railway in some capacity, with named customers including Bilt, Intuit’s GoCo subsidiary, TripAdvisor’s Cruise Critic, and MGM Resorts. Kernel, a Y Combinator-backed AI infrastructure startup serving more than 1,000 companies, reportedly runs its customer-facing systems on Railway for $444 per month, according to VentureBeat. For enterprise buyers, VentureBeat notes Railway offers SOC 2 Type 2 compliance, HIPAA readiness, single sign-on, audit logging, and a “bring your own cloud” deployment option, with add-ons such as extended log retention priced at $200 monthly and HIPAA business associate agreements at $1,000.
Hype Check
Claim: Railway says its AI-native infrastructure lets it out-deploy and undercut legacy cloud giants like AWS and Google Cloud, delivering sub-one-second deployments and cost savings of up to 65 percent, per VentureBeat. Reality: VentureBeat’s reporting shows genuine enterprise traction, verifiable customer figures like G2X’s cost drop from $15,000 to about $1,000 monthly, and a lean 30-person team generating tens of millions in revenue, but these are largely company- and customer-supplied metrics rather than independently audited benchmarks, and Railway remains far smaller in scale than AWS or Google Cloud. Verdict: Mixed — the funding, customer base and technical differentiation are substantive, but the comparison to hyperscaler-level dominance is still aspirational. This is not financial advice.
Source
Researched with AI assistance, fact-checked and edited by a human. Not financial advice.