For those of you who have been following Gremlin since 2017, you know that they have long advocated for engineers to be more proactive about the reliability of their systems. The founder Kolton Andrus cut his teeth at Amazon and Netflix, and did some amazing work to make Chaos Engineering a known practice to software engineers worldwide.
The premise of Chaos Engineering is straightforward: by putting your system through some upfront tests, SREs and platform teams alike can find where the system may be brittle, before that crack has the chance to burst open and cause real customer pain. But the reality on the ground is that most businesses do not have the engineering resources of a Netflix or an Amazon; engineering teams seem too busy fighting fires to actually set aside time to do necessary proactive reliability work.
But it looks like that’s about to change.
Gremlin has just made Foresight AI generally available. It’s designed to automatically spot potential system failures, recommend fixes, and test whether those fixes actually resolved the risk. Put simply: engineering teams now have little excuse to not practice proactive reliability. As Andrus puts it, this is like being able to go to the gym in your sleep. Who wouldn’t want to have healthier and more reliable systems?
“AI-driven development means shipping code at 10X velocity; it also means 10x the opportunity for bugs, risks, and failures,” said Kolton Andrus, CEO and Founder of Gremlin. “While the rise of AI SRE tools are great for helping teams respond to incidents faster, it’s still cleanup after something breaks. Gremlin Foresight AI finds risks proactively, delivers the fix, and verifies the fix worked by rerunning the test. This provides teams with the assurances needed to move confidently.”
The product draws on Gremlin’s Failure Atlas, a collection of more than a decade of data about how online systems fail and recover. Gremlin says this allows Foresight AI to base recommendations on observed failure patterns rather than generic guidance.
The system can identify potential weaknesses, propose remediation as a configuration patch or infrastructure-as-code change, and rerun the test that exposed the risk to check whether the fix worked. It also tracks reliability scores across services and teams, which Gremlin says can help organizations measure progress and decide where to invest.
Foresight AI follows a beta program that Gremlin says demonstrated the product’s ability to detect and address risks before they become incidents. The company describes the tool as an extension of its broader reliability platform, which supports planned experiments, limits their potential impact and automates testing over time.
Gremlin was founded by engineers with experience at Amazon and Netflix. Andrus previously worked on uptime for Amazon’s retail site and later helped build fault-injection tools at Netflix, following the release of the open-source project Chaos Monkey.
