Yonas Beshawred
@yonasbe
Co-Founder @starslingdev | Prev. Founder & CEO @stackshareio (acq)
1.1K Following 2.1K Followers
Today we’re announcing our pre-seed round and a new product: @starslingdev Review Runners - code review agents that use your model key, your team’s skills and scripts, and reviewers you define, all running in GitHub Actions.
We’ve raised $3M from @BessemerVP and @ycombinator to make code verification agent-native.
code → verification (ci + code review) → production
Show more
Announcing StarSling GPU Runners: GitHub Actions runners backed by fast NVIDIA GPUs ⚡️
More and more teams need GPUs for their CI, and those GPUs have to be pulled from training runs or inference. Your GPUs should be used for product, not tests. With StarSling GPU Runners you get dedicated GPUs for your CI jobs with per minute billing.
One of our customers is already seeing a 14x reduction in queue time for a GPU runner!
Show more
CI Harness v2 - an agent loop that thinks like a performance engineer 💫
Our first production run produced four merged pull requests into Better Auth, including one test job that went from 24 seconds to 4.5s.
Built on @mastra @ClickHouseDB @hatchet_dev @render ✨
Show more
We wrote about how we rebuilt our CI agent loop to think more like a performance engineer.
End result: @better_auth got a 5X on one of their jobs⚡️
PR:
Details on how we got agents to think like a real performance engineer to find CI speedup wins 🏎️
The first 80% of vibe coding feels like magic.
The last 20% teaches you why software engineering exists.
Prove me wrong?
Our CLI stack:
- @bunjavascript: compiles it to one static binary
- Ink (@vadimdemedes): the TUI
- @better_auth: social sign-in and device-code
- @arktypeio: end-to-end typesafety
API stack:
- @elysiaJS (@saltyAom): the routes, published as OpenAPI
- @nitrojsdev: packages and deploys the server
- @vercel: hosting
- @PlanetScale + @DrizzleORM: Postgres for identity, orgs, API keys
- @upstash: session lookups that skip the db
- @ClickHouseDB: every runs / jobs / top / usage query
- @github: identity and raw job log ingest
- @stripe: sling bill
Show more
Today we shipped sling, an agent-first CLI for your GitHub Actions 💫
Right now, if you ask your agent why a CI job failed, it goes and pulls the entire set of GHA logs and eats up your context window.
🪵 > sling why @starslingdev computes the run's logs and metadata, along with the evidence lines it matched to tell you why a run failed
If you want to know what your GHA usage is, you have to go to the GH UI and click around to find it.
💸 > sling bill tells you how many minutes, how much you're spending, and across which runner labels
Trying to understand what your slowest job is? You'll need to hunt around the GHA UI and even then you only get an average.
📊 > sling top and sling usage tell you runner minutes and cost across all workflows and repos, with p50, p95, p99, queue wait and a week over week trend
Show more
All credit to @binaryshrey & @dbworku!
This is slick. Expect nothing less from @yonasbe.
This is slick. Expect nothing less from @yonasbe.
Today we shipped sling, an agent-first CLI for your GitHub Actions 💫
Right now, if you ask your agent why a CI job failed, it goes and pulls the entire set of GHA logs and eats up your context window.
🪵 > sling why @starslingdev computes the run's logs and metadata, along with the evidence lines it matched to tell you why a run failed
If you want to know what your GHA usage is, you have to go to the GH UI and click around to find it.
💸 > sling bill tells you how many minutes, how much you're spending, and across which runner labels
Trying to understand what your slowest job is? You'll need to hunt around the GHA UI and even then you only get an average.
📊 > sling top and sling usage tell you runner minutes and cost across all workflows and repos, with p50, p95, p99, queue wait and a week over week trend
Show more
💫 sling - the agent-first CLI for GitHub Actions
AI infra is the other big winner. No one wants to be on the hook for the availability of complex critical path services. These aren’t the ones you can build with Claude. They’re the kind that power it:
Show more
AI-native services is the platform shift of this era. No one wants another SaaS tool that they can build with Claude in a day.
Ran hpc-sandbox-benchmarks on @cursor_ai cloud agents...
They give you:
- 4 vCPU, 16GB RAM with 270GB disk
- Intel Emerald Rapids chip
- AWS us-east-2 (AS16509)
- 4.2 Mbit/s up and down
Show more
3 @github Actions skills shipped in the past month, more on the way 💫
Our third agent skill is live! Check your repo's GitHub Actions for 10 attack vectors that all map to public security incidents.
npx skills add starslingdev/skills
Easily check your GitHub Actions against 10 attack vectors all backed by a named public incident or published security research with /ci-secure 🔐
Have you noticed that people have stopped saying that software engineering is over?
worktrees must die
this is 20GB of repeated node_modules lol
Thread on where each sandbox provider shines 👇
The latest version of hpc-sandbox-benchmarks is live!
We’ve added 5 additional sandbox providers: @microsandbox @namespace @RunloopAI and @vercel
Today we’re also releasing an interactive results explorer
that allows you to cut the data across the benchmarks to see p50, p95, and specs for CPU, disk I/O, memory, network, etc.
Try out the explorer, we’d love your feedback! And keep the PRs coming :)
Show more
StarSling Runners are now in GA! We did a ton of infra work to make our runners self-serve and are super excited that anyone can try them out now.
One customer who works with the frontier AI labs just signed up and migrated two of their most problematic workflows over to StarSling and saw an immediate 50% speedup in their jobs 🏎️
Every team should have faster, smarter, and cheaper CI.
Show more
