Railway just raised $100M — one client cut AWS bills 87%

If you've ever tried to deploy an app to Amazon Web Services (AWS) and spent hours buried in configuration screens before anything actually ran, Railway wants to fix that — and $100 million in fresh funding says it might succeed.

Railway, a cloud platform (a service that runs your apps and websites on the internet) founded by 28-year-old Jake Cooper in 2020, just closed a $100M Series B round led by TQ Ventures, with participation from FPV Ventures, Redpoint, and Unusual Ventures. That brings total raised to $124M.

Who's Actually Using This?

Railway has quietly become a serious platform. Current numbers:

• 2 million developers globally, growing at ~200,000 new users per month
• 31% of Fortune 500 companies use Railway for some workloads
• 10+ million deployments monthly
• 1 trillion+ requests handled through its edge network (the global network of servers that handles incoming traffic)

Customers include TripAdvisor, MGM Resorts, and thousands of AI-native startups. The company is generating $30M+ in annual recurring revenue on a team of just 30 people, growing at 15% month-over-month.

The 87% AWS Bill Cut — Is It Real?

Daniel Lobaton, CTO at G2X (a platform serving 100,000 federal government contractors), reported cutting infrastructure costs from $15,000/month down to ~$1,000/month after migrating to Railway — an 87% reduction. Deployment speed also improved 7×.

That's an extreme case, and Railway itself claims more modest savings: roughly 50% cheaper than major hyperscalers like AWS and Azure. But the mechanism is real. Railway bills by the second on actual CPU (computing power) and memory you use, rather than charging you for a fixed server whether it's busy or idle. For most startups and AI applications, you're paying for peak capacity 24/7 on AWS but actually using it 20% of the time.

What Makes It "AI-Native"?

Cooper, who previously worked as a software engineer at Wolfram Alpha, Bloomberg, and Uber before founding Railway at age 22, says the old cloud platforms were designed for a pre-AI world. Railway was rebuilt from the ground up for 2024-2026:

• Deployments under 1 second — vs. 2–3 minutes via traditional tools like AWS CodeDeploy
• MCP server integration (added August 2025) — MCP, or Model Context Protocol, lets AI tools like Claude directly hook into your Railway infrastructure and control deployments automatically
• Claude integration — Railway has a direct Claude connection that lets AI agents "analyze infrastructure, loop deployments, and auto-fix errors" without a human in the loop
• "Agentic primitives" — building blocks designed to let AI agents move 1,000× faster through deployment cycles

Who Backs This?

Beyond institutional investors (TQ Ventures, Redpoint, FPV Ventures), Railway's angel investors read like a who's-who of developer tools:

• Tom Preston-Werner — co-founder of GitHub
• Guillermo Rauch — CEO of Vercel (the platform behind Next.js)
• Olivier Pomel — CEO of Datadog
• Spencer Kimball — CEO of Cockroach Labs
• Jori Lallo — co-founder of Linear

These aren't passive investors. They're the people who built the tools that developers use every day, betting that Railway is where those developers go next.

How to Try It

Railway has a free tier that includes $1 in resource credits per month — enough to run a simple app and get a feel for the platform. Paid plans start at $5/month (Hobby) for personal projects, or $20/month (Pro) for production apps. Enterprise pricing is custom.

Get started at railway.com — deploy a project in under a minute with no configuration required.

Related Content — Get Started with Easy Claude Code | Free Learning Guides | More AI News