How G2X Unlocked Rapid Experimentation at Scale with Railway

For a go-to-market platform serving 100,000 federal contractors monthly, every deployment matters. G2X provides market intelligence and business productivity tools to companies selling to the federal government, processing 3 to 5 million requests daily while ingesting data from two dozen notoriously unwieldy government sources that share opportunities totaling billions of dollars of RFPs.

As G2X's engineering team, led by CTO Daniel Lobaton tried to scale their platform on a major public cloud provider, infrastructure complexity became a massive bottleneck to innovation.

"The work that used to take me a week on our previous infrastructure, I can do in Railway in like a day. Just spinning up a database instance, waiting for the thing to actually come up can easily take like 20 minutes."

• Daniel Lobaton, CTO at G2X

The complexity went beyond just deployment speed. Every new service required navigating virtual private clouds, security groups, network address translation, and permission systems. For a lean engineering team trying to iterate quickly on product features, the infrastructure overhead was crushing.

"If I want to spin up a new service and test different architectures, it would take so long. What do you do in the meantime? You go out, make coffee, draft an email... In Railway I can launch like six services in two minutes.

Cost was equally problematic. G2X's cloud infrastructure bill had ballooned to $15,000 per month at its peak, with significant waste from unused services and over-provisioned resources that were difficult to identify and clean up.

This all changed when G2X began the migration to Railway.

G2X discovered Railway through their consulting partner Kirana, who positioned it as a rapid prototyping platform. What started as an experiment quickly became their entire production infrastructure.

The migration was remarkably straightforward. Railway's Git-based deployment model meant G2X could maintain their existing development workflow without changes. Furthermore, the platform's PR environments provided instant staging for every pull request.

"Railway practically just follows whatever branch strategy you got. Just push that branch and it's there. Then, the main benefit of PR environments is being able to click on the link from the PR and then check the application. I absolutely love that."

Railway transformed how quickly G2X could experiment with architectures. The platform's simplicity eliminated the networking complexity that plagued their previous setup.

Migration was simple, the G2X engineering team just had to point Railway to their git repo, expose the app internally, set up a tunnel and they were go to go.

"You don't have to deal with messy networking rules. You don't have to check the NAT gateway or check the security rule. With Railway, if you understand basic TCP/IP, you got it down."

A critical confidence boost came when Railway migrated off Google Cloud Platform to their own infrastructure.

"When we migrated, reliability massively shot up. We felt much more confident about our platform, so I brought in even more critical services then. We were no longer worried about if GCP was down - Railway just dealt with it."

The simplicity extended beyond just deployment. Railway made it easy to spin up supporting infrastructure like databases and caching layers without complex configuration, especially as compared to traditional cloud providers.

Railway's impact on G2X was transformative, affecting both their bottom line and their ability to innovate. For a platform serving the complex, mission-critical needs of federal contractors and the F500, Railway eliminated the traditional tradeoff between simplicity and power—delivering both in a package that costs 87% less than their previous setup while enabling the team to move seven times faster.

• Infrastructure costs plummeted 87%, from $15,000 monthly on their previous cloud provider to approximately $1,000 on Railway. This dramatic reduction came while actually improving performance and reliability.
• Deployment velocity increased 7x. What previously took a week now takes a day on Railway. Environment creation that required half a day of configuration now takes minutes.
• 100,000 monthly active users served reliably with 3-5 million daily requests processed smoothly. During peak hours, their most demanding service handles 250 requests per minute without issue.
• Engineering focus shifted from infrastructure to product. The simplicity freed the team to work on what matters: serving federal contractors better.
• Reduction of most demanding load by 75%, using aggressive caching to reduce RPM by 3-4x, down from 1,000 RPM.

Looking forward, G2X sees Railway as their primary infrastructure platform indefinitely. G2X is part of the daily workflow of thousands of users, many of which find G2X mission critical.

The platform handles G2X's complex data pipeline entirely- ingesting, cleaning, and serving data from two dozen federal sources, with the largest approaching 1 terabyte of raw text.

"We're dealing with several external data sources that you have absolutely no control over, most of them being very unreliable," Lobaton explains. "How do you make that which is naturally unreliable, reliable? That's what underpins G2X."

For a platform serving the complex, mission-critical needs of federal contractors, Railway eliminated the traditional tradeoff between simplicity and power- delivering both in a package that costs 87% less than their previous setup while enabling the team to move seven times faster.

If you're interested in optimizing your infrastructure and developer velocity at scale, reach out to us: railway.com/enterprise.