OpenAI's CI playbook (for GitHub Actions)
OpenAI's Core Experimentation team cut median CI time from about 10 minutes to 4.5. Ten of its techniques, each with a copy-paste prompt for your agent.
·Yonas Beshawred
The StarSling Blog
Anthropic's CI playbook, as GitHub Actions prompts for your agent
Anthropic's CI jobs grew 25x in six months. How to run only the tests a change affects in GitHub Actions, with a copy-paste prompt for your agent per technique.
·Yonas Beshawred
Introducing Adaptive Compute: up to 2.3x faster GitHub Actions
StarSling Runners with Adaptive Compute ran up to 2.32x faster than GitHub-hosted runners, boosting CPU up to 25% and RAM up to 50% at no extra cost.
·Daniel Worku
Linear's CI playbook, ready to hand to your agent
Linear's 15 CI changes from their post, each with a copy-paste prompt that hands the move to your coding agent and names the ci-speedup pattern behind it.
·Yonas Beshawred
Ubuntu 26 generally available
GitHub moves ubuntu-latest to Ubuntu 26.04 starting Oct 19, 2026, and the switch can break builds. Use our agent prompt to check if your jobs are affected.
·Yonas Beshawred
StarSling tops the GitHub Actions runner benchmarks
StarSling Runners finished first on the Rust and TypeScript builds in RunsOn's public GitHub Actions runner benchmark, across 9 providers and 89 runners.
·Yonas Beshawred
Levels of CI
Five levels of CI optimization, from the Actions cache to swapping runners, tuning workflows, switching CI engines, and replacing CI with agents.
·Yonas Beshawred
Announcing Our $3M Pre-Seed Round and StarSling Review Runners
A $3M pre-seed from Bessemer Venture Partners, Y Combinator and five more funds to make shipping code agent-native, and Review Runners in private beta.
·Yonas Beshawred
Announcing StarSling GPU Runners
GitHub Actions runners backed by NVIDIA GPUs: RTX PRO 6000, RTX 5090 and RTX 4090, with H100 coming. One line in runs-on. Contact us to get started.
·Yonas Beshawred
The agent loop that shipped a 5.4x speedup
The job was green. The logs were clean. CI Harness v2 found the 11-month-old bottleneck anyway, and made that job 5.4x faster.
·Yunjia Zheng