Komodor wants to make every engineer a Kubernetes troubleshooting expert

For a time, Kubernetes, the open source container management platform, was as hot as it comes, but lately it has settled into a more mainstream cadence of a maturing technology. As more companies adopt containerization and micro services, it requires more sophisticated tooling to manage a system with loads of underlying complexity.

Komodor’s two founders cut their teeth at Google and eBay working on these types of systems, and they experienced what many folks at organizations with large engineering tems often encounter. While these big companies had the resources to build tooling internally to manage these systems, other companies were forced to do things more manually.

Two years ago, Ben Ofiri and Itiel Shwartz left the comfort of their corporate jobs to start  Komodore, and build a Kubernetes troubleshooting platform, one that could help every organization find and fix issues in Kubernetes installations.

“Once companies start to adopt micro services and Kubernetes, they are all facing the same challenges and issues. Kubernetes is a very, very complex system, very distributed, very fragmented, and is actually composed of thousands of different components,” Ofiri, who is the startup’s CEO, told me.

He says that when an incident happens, there is a lot of pressure on the engineering team to figure out the nature of the problem and fix it as quickly as possible. The trouble is that most people aren’t adequately trained to deal with these issues, he said.

He said his company wanted to put that troubleshooting capability into more engineer’s hands using software to help. “What we have tried to do at Komodor is to democratize the operational and troubleshooting aspects of Kubernetes and take this knowledge that maybe a few people in the organization have, and to expose it to the other 95% of the organization,” Ofiri explained.

This involves detection, investigation and remediation. “What we do behind the scenes is we leverage different data-driven approaches and a rule engine-based model in order to first of identify different issues, and then to come up with proposals on how to automate the investigation phase in order to find the root cause.”

They launched the company in 2020, came up with the first draft of the solution about a half year later with beta customers. They have had a full-fledged solution in production for almost a year now. The company already has 45 employees, and he says one way he has been able to hire people from diverse backgrounds is by training people who didn’t have direct experience as developers.

“We both take diversity and inclusion very seriously, and we’re making sure that [a variety of] people are getting opportunities at Komodor. We hired entry level people without any programming experience,” he said.

While it was challenging to train developers, now that they’ve got a program in place, it’s a much smoother process. He says that while it’s popular for startups to avoid an office experience these days, he still sees a lot of value in working together in the same building, and he’s hoping to be able to provide that experience as the company grows.

Today the company announced a $42 million Series B investment led by Tiger Global with participation from Felicis and existing investors Accel, NFX Capital, OldSlip Group, Pitango and First and Vine Ventures. The startup has raised a total of $67 million.