Sequoia-incubated Empirik launches with $21M to predict outages before they happen
TechCrunch Marina Temkin
Sequoia-backed Empirik launched with $21M to spot outages before they hit. It watches system changes and flags risky ones, so engineers don’t have to play cleanup crew all day.
Based on reporting by TechCrunch, Marina Temkin — read the original for the full story.
Summary, retelling and take written by AI under human oversight; images are AI-generated illustrations. How we work · Report an error
Empirik is trying to make outages less of an after-the-fact problem and more of a thing software catches before anyone feels it. The startup came out of Sequoia’s own shop, then announced Tuesday that it has spun out as an independent company with $21 million in seed money from Sequoia, Canapi, and Alumni Ventures.
The idea started with Avon Puri, who spent more than a decade running infrastructure at Rubrik and VMware before joining Sequoia in 2020 as chief digital and information officer. About three years ago, as large language models started showing what they could do, Puri and Sequoia IT leader Sudheer Dhurjati saw a gap: infrastructure teams still spend too much time reacting after systems break. Empirik was built to track changes across a system and guess how those changes might ripple outward.
Sequoia incubated the company in 2023 and later brought in Kartik Chandrayana as CEO earlier this year. Chandrayana came from Quantum Metric, where he was chief product officer, and from Salesforce, where he worked on observability. The pitch is simple enough: let the software act like an autonomous traffic cop, approving low-risk changes, adding guardrails around bigger ones, and kicking the worst cases to a human.
That’s not just a nice demo. Empirik says it already has customers, from startups to several Fortune 500 companies, including S&P Global, Guardant Health, and a major consumer packaged goods company. Sequoia partner Bogomil Balkansky says most observability tools still miss the web of dependencies inside modern infrastructure, and argues Empirik is one of the first tools built specifically to track changes across huge environments. The company wants to do for infrastructure engineers what Cursor and Claude Code did for software developers: automate the grunt work so people can move faster.
For now, Sequoia says Empirik sits in a category of its own, though Balkansky places it alongside AI SRE platforms like Resolve and Sequoia-backed Traversal as a complementary layer. That’s a neat way of saying the market is still being drawn in pencil, not ink. But the problem is real, the budget is there, and infrastructure teams are clearly tired of being the ones who find out something broke after everyone else does.
My take — AI-written commentary, not fact-checked reporting
This is the right kind of AI startup: narrow, painful, and attached to an actual budget line instead of a demo reel. The industry keeps pretending observability is solved, then another outage reminds everyone that logs aren’t the same thing as understanding. And yes, the closed, enterprise-heavy route makes more sense here than a shiny open model free-for-all; nobody wants their outage predictor to be crowd-sourced by optimism.
Read more about this at: TechCrunch
