Gremlin Chaos Engineering Platform

Gremlin is a chaos engineering platform helping organizations improve system reliability by proactively identifying and fixing vulnerabilities through controlled failure experiments. Founded in 2016 by Kolton Andrus (CEO, former chaos engineer at Amazon and Netflix) and Matthew Fornaciari in San Jose, California. Raised $33 million from Redpoint Ventures, Index Ventures, Amplify Partners. Serves Amazon, Netflix, Twilio, Capital One, National Australia Bank. First company to commercialize chaos engineering, building on the Chaos Monkey concept from Netflix. Key features: chaos attacks: safely inject controlled failures into infrastructure. Attack types include: CPU stress, memory stress, disk stress, IO stress, network latency, packet loss, DNS blackout, process killer, shutdown (reboot machines), time travel (change system clock), blackhole (drop all traffic to a host or service), HTTP request modification. Scenarios: multi-step orchestrated chaos experiments. Define sequences of attacks with blast radius, impact targets, and halt conditions. Run complex failure simulations mimicking real-world outages. Blast radius control: start small (single container) and gradually expand (full cluster). Target specific hosts, services, regions. Safeguards: automatic abort on health check failures. Custom halt conditions based on metrics from Datadog, Prometheus, New Relic. Time-limited runs. Manual abort. Kubernetes support: target pods, deployments, daemon sets, nodes. Alfi: application-level chaos for mobile and web apps. Game Days. Integrations: Kubernetes, Docker, AWS, Azure, GCP, Datadog, Prometheus, PagerDuty, Slack. Go/React. Proprietary.