What problem does it solve?
Autosearch eliminates guesswork in Kubernetes and OpenShift performance tuning by running measured experiment loops, comparing results against a baseline, and keeping only changes that produce real improvement.
Core Features & Use Cases
- Continuous Optimization: Iteratively tune CPU, memory, replicas, rollout behavior, build speed, and latency until a target is reached.
- Session Persistence: Preserve experiment history across restarts with log files and a living progress document.
- Operational Safety: Record outcomes, revert bad changes, and validate results with health checks and rollout measurements.
- Use Case: A platform engineer can reduce a service’s memory footprint by testing one resource change at a time, measuring the impact, and committing the winning configuration.
Quick Start
Use the autosearch skill to optimize a Kubernetes service by defining a metric, establishing a baseline, and iterating on one change at a time.