Why we couldn't resist raising from eBay - our Series A announcement

This is not your typical 'we raised X amount of money' post, so let's get the boring part out of the way first, or read the official announcement in TechCrunch .

At the beginning of 2025, WunderGraph was already profitable. With a staggering burn multiple of 0.14, we were able to end 2024 with virtually infinite runway, so why raise? This is where eBay comes into the picture, but let's start from the beginning because our relationship with eBay started way before 2025.

It was summer 2023, and WunderGraph was not doing well. The "Backend For Frontend" Framework we built was not gaining the traction we had hoped for. The whole team was together on a retreat in an old sheep farm in the Netherlands, realizing that we had to pivot.

We had built an API Gateway capable of turning any API into GraphQL, including rudimentary support for Apollo Federation. As we looked at our notes from sales calls, we concluded that there's an opportunity right in front of us to turn the ship around. Side note: These were handwritten notes; we didn't have AI note-taking apps back then.

The decision was made to build a new product, a fully open-source Platform for GraphQL Federation, and that's how Cosmo was born. Two months later, we onboarded our first customer, but before Cosmo took off, we had to go through the darkest days in our company's history.

Cosmo was barely working, and we were running out of money. We quickly needed to ramp up our sales efforts to survive. Our last hope was the GraphQL Conf in San Francisco, where we presented Cosmo for the first time to the public.

What not many people know is that while I was on stage, holding a talk about Dataloader 3.0 , we kind of knew that we weren't going to make it. We were still optimistic, but it was clear that a miracle was needed.

It was almost cinematic when 10 people from Apollo were suddenly standing in front of our tiny booth. They probably made jokes about us, or at least they did not take us seriously. If only they knew that we only had three customers and 3k MRR at the time. It was really hard to stay positive, but we were not yet done.

That said, our trip to San Francisco was not a complete loss. We met some amazing people and built important connections that would soon change the trajectory of our company.

One memory that sticks with us from that trip is when we met two amazing gentlemen in their golden years at Benihana. We were lucky to share a table with Ale and Bill and couldn't resist asking them for a picture. Two brothers, both in their 90s, living their best lives while traveling the world. This evening made us realize that there's more to life than just building a startup; the relationships we build along the way make it all worth it.

When we got back from San Francisco, we had more than 30 scheduled sales calls, all of which came from the conference. However, initial enthusiasm quickly turned into disappointment when we realized that the initial interest was not translating into actual customers.

Before the end of 2024, we could only count two customers on Cosmo, accounting for 2.5k MRR. One of our first customers got poached by a competitor during GraphQL Conf, so we lost them. At the time, we were burning through our runway quite fast, with only 6 months left.

One of our biggest challenges at that time was not just a lack of pipeline but also issues with our implementation. One of the critical parts of Cosmo is the Query Planner, which is responsible for generating a query plan to load the data from the underlying APIs. It was tricky to get this part right, as Federation has lots of edge cases, including some that are not even known to the creators. Luckily, our sales cycles were long enough that we could land more customers, just in time.

While the first two months of 2024 were almost dead, we slowly picked up momentum in March and landed our first bigger customers. Then, suddenly, the unthinkable happened.

It was early March 2024, and we were expecting a regular day when suddenly we got a notification in our #general channel in Slack about a new customer that had signed up for Cosmo. We've set up a Stripe integration to boost motivation when a new customer signs up. Our self signup plan is just $500/month, but every customer counts.

It wasn't the first time the notification popped up, but this time it was different. Why would someone from eBay use the self signup plan? Why are they not using our "talk to sales" form? Is this real, or a scam?

We reached out to them, and shortly thereafter they joined a private channel on our Discord server. What started as a lengthy conversation about the capabilities of Cosmo quickly turned into a discussion about our vision for the future of GraphQL. Eventually, we met via Zoom and it became clear that our values and vision in terms of how GraphQL should be used in enterprise matched. From that point on, it took no more than 2 months to finalize and sign the contract.

Here's what Bryan Woodruff, VP of Seller Experience Engineering at eBay said about us:

What followed after March was absolute madness. We were suddenly overwhelmed with demand and had to onboard new customers. Among them were big names like On The Beach, SoundCloud, EOG Resources, INNIO, and many more super cool companies (unfortunately under NDA, so I can't mention them). We were heads down for the rest of the year and barely had time to breathe. You may have noticed that for quite some time we went quiet on our blog and social media channels.

Towards the end of 2024, we suddenly woke up to the fact that we were profitable. It was time to think about the future and how we were going to scale.

Fast forward to early 2025, the founders of WunderGraph had to make a difficult decision. Are we going to keep bootstrapping or raise a Series A? Fortunately, eBay gave us a little push in the right direction.

We thought to ourselves, if we can make it work for eBay, we can make it work for everyone. Thanks to their experience in scaling GraphQL and the sheer size of their Federation deployment, we knew that this was a once-in-a-lifetime opportunity.

Together, we're aiming to build the Enterprise standard for GraphQL Federation. A fully modular open-source platform that is flexible enough to be used in any environment but also has the capabilities to be deployed at the scale of eBay.

Every story has some twists and turns, so I'd like to share two stories that made a big difference in our journey. A big enough difference that I'm not sure we would be here today without them.

Shortly after our retreat in the Netherlands, we were approached by Tailor Inc., a YC backed startup that builds a headless ERP system. At the very core of their architecture is a GraphQL Federation gateway that is powered by the GraphQL engine of WunderGraph, a library we've been working on since 2018, which also powers Cosmo Router.

The guys from Tailor asked us if we could implement Federation v2 support in the engine. Their commitment was a key factor to not just fund a part of the development of the engine, but also help us with use cases and feedback to drive the project forward. At this point, we'd like to personally thank Yo Shibata (CEO) and Misato Takahashi (CTO) for their trust and belief in us. We would also like to thank Jack Chuka (Head of Platform), our main point of contact at Tailor.

The second story is about Sven and Fredrik from Soundtrack Your Brand . Shortly after launching Cosmo, Fredrik and Sven reached out to us, showing interest in using Cosmo for their internal GraphQL Platform. Their use of GraphQL Federation was very advanced, too advanced for us to handle at the time. They knew that we had to do a lot of work to get there, but they liked the vision and wanted to be part of it.

Having access to their schemas, queries, and use cases was a game changer for us. Betting on us pushed us to work day and night to make it happen. It took us until the end of January 2024 to get them into production, but we did it, and from that point on, growth was exponential.

Dear Fredrik and Sven, thank you for taking a chance on us. Our massive growth would not have been possible without you pushing us to the limit in the early days.

Aside from partnering with a publicly traded company, we're excited to welcome Tommi Uhari, Partner at Karma Ventures, to our board of directors, representing our new lead investor.

Karma Ventures is a very tech-savvy VC firm from Tallinn, Estonia, that focuses on deep tech software companies. Over an in-person dinner, we concluded that our visions align well in terms of how we see the future of APIs, AI, and the role of GraphQL in the enterprise.

We're extremely proud of what our team of just under 20 people has achieved in the last 12 months, but our journey is really just at the beginning. We've got exciting plans for the future, and we're looking for more people to join us.

You've probably heard the typical "we're only getting started" speech, but seriously, that's exactly where we currently are. We've figured out how to make GraphQL Federation work for small startups and big companies like eBay, we've got a go to market strategy that works, and now we're ready to scale.

On the product side, it's long overdue to build a better version of Federation. While Federation v1 built a great foundation, v2 introduced more problems than it solved. Our goal is to help Federation get back to its roots, making it stricter and simpler, but also open to any API, not just GraphQL. In addition, we're working on a new product to help developers discover & share APIs more easily while enabling better workflows for collaborating on APIs.

To make all of this happen, we're looking to expand our team on the sales, marketing, and engineering side.

If you're an experienced sales leader, we're looking for you to join our team as VP of Sales. So far, we've been very successful growing our customer base with founder-led sales efforts. Now, we're looking for a proven sales leader to help us scale our go-to-market efforts.

On the marketing side, we're looking for a VP of Marketing to help us reach more people and grow our community. If you're a marketing expert, you can join this rocket ship and help us build and expand our Enterprise funnels.

In addition to sales and marketing, we're also looking for Software Engineers in all areas.

If you're passionate about Golang, APIs, and Open Source, you're a perfect fit to help us build the future of GraphQL Federation. At the core of WunderGraph is the GraphQL engine and Cosmo Router. This is a unique opportunity to build API Gateway infrastructure for companies like eBay, On The Beach, SoundCloud, and many more. We have plans to build a better version of Federation which is not just limited to GraphQL, but open to any API, Kafka, OpenAPI, gRPC, and more.

On the frontend & fullstack side, we're investing heavily in the developer experience of APIs. Our goal was always to build the "GitHub for APIs", a place where developers can easily discover APIs, understand their capabilities, share and collaborate on APIs. We're now making this a reality and you can be part of it.

At the heart of WunderGraph is the belief that every digital business runs on top of APIs. Apps and AI-powered services, as well as agents, need access to real-time data to be truly useful.

Our goal is to help organizations unify all their APIs into one place, making it easy to discover APIs for their use cases and collaborate on creating new APIs for future products.