The data architecture of today's enterprises is completely different from 20 years ago, but monitoring and tracing tools haven't kept up, making it difficult for engineering teams to pinpoint the root cause of errors or latency issues.
Israeli developer Eden Federman believes the solution is a more advanced observability method: distributed tracing, but has found that adoption is lower than he expected due to high implementation hurdles. This prompted him to co-found Odigos, a startup that helps companies easily monitor the complex, cloud-native systems that process massive amounts of data, as is often the case today.
He and CEO Ari Recht are not related, but their wives are cousins. So Federman asked Recht (standing left in the photo above), an investment banker, to pitch him the idea. This quickly turned into a series of meetings. “The more we worked together, the more I liked him and the more I realized he was missing the business side of things,” Recht told TechCrunch.
Unlike Federman, Recht doesn't code, but he's spent a lot of time in startups and is clearly used to being the business face of tech-focused Odigo. They built a compelling enough team that Y Combinator selected them to join its Winter 2023 batch, which is mostly in-person in San Francisco.
“We actually lived there together for six months, and it was a very fulfilling time for us,” Recht said. One highlight was meeting the angel investors who eventually backed Odigo. The company, originally called Kibal, raised money through safe notes around the time of YCombinat's Demo Day and more recently in a $13 million round led by Boston-based venture capital firm VentureGuide.
One of the angel investors in this round, Lightstep co-founder Ben Sigelman, is also the co-creator of OpenTelemetry, an open-source observability framework that Odigos uses and helps maintain. Recht called it “amazing,” but also said it's “very difficult” to implement and configure. That gives Odigos an opportunity to offer “enterprise-grade OpenTelemetry” to its customers.
Odigos' target market is large enterprises – not the Googles of the world, not the small startups. Large enterprises can usually solve distributed tracing on their own, including with Odigos' open source repositories. But less tech-savvy enterprise clients need help implementing what would otherwise require a huge amount of manual work.
This is where eBPF comes in. It stands for Extended Berkeley Packet Filter, and is a technology that allows developers to write programs that run directly on the Linux operating system. Different startups use it in different ways, but for Odigos, it was the key to enabling automatic distributed tracing without undue impact on performance or requiring code changes.
This is especially valuable for enterprise companies, Recht says: “If you have 25,000 applications, or 25,000 microservices, doing it manually is nearly impossible. And doing it manually creates a lot of performance overhead.”
Federman and most of the Odigo team members are active open source contributors, especially in the eBPF and OpenTelemetry groups, and the startup donated its first project to the Cloud Native Computing Foundation, which also helps it understand what its enterprise clients are missing. “I think one of the reasons customers trust us is because of our open source background,” Federman says.
As is common with startups that have an open source side, the commercial version of Odigos isn't exactly the same: Federman says the open source version of Odigos works very well with a single Kubernetes cluster, but the enterprise edition supports multiple environments.
More importantly, Odigos offers large organizations the opportunity to implement distributed tracing without leveraging developer resources or becoming experts. Odigos also does not require them to change their application performance monitoring tools, such as Datadog, Dynatrace, SigNoz, or Honeycomb. Honeycomb CEO Christine Yen invested in Odigos' latest round.
Others who participated in the latest round include Firestreak Ventures, Mango Capital, Salesforce Ventures, and various angel investors. The new funding will help the startup develop its solution, invest in a go-to-market strategy, and set up a new headquarters in Boston, where the lead investor is based and where the founders plan to eventually relocate. This will bring the company closer to its prospects. Though still in stealth mode, the company recently signed a “leading proof of concept” with a very large US company.