
Managing Partner @precursorvc. Fan of @chelseafc @lafc @lions and all things Michigan. Most days in Jackson Square in SF. I invest $250-500K at pre-seed or seed
San Francisco, CA
Joined February 2007
- Tweets8.7K
- Following3.7K
- Followers46K
- Likes8.6K
Charles Hudson retweeted
You can now advertise directly to people’s agents with ReachAgents.
Instead of pursuing GEO, send agents your whole brand and product information.
Each agent then shows its owner a unique ad, in their feed, at the moment of intent.
Launching today at ReachAgents.com/advertisers
Charles Hudson retweeted
Levels of CI - a full guide for making your GitHub Actions cheaper and faster 👇🏾
Level 1: use Actions cache
cache every dependency install, keyed on the lockfile so it actually hits.
🔗 Docs: docs.github.com/en/actions/r…
🔗 Agent prompt: starsling.dev/best-practices…
Level 2: swap runners
get faster, cheaper machines by changing the runs-on label. Your workflows, triggers, and checks stay exactly as they are.
🔗 Runner comparison: starsling.dev/compare
🔗 Right-size runners: starsling.dev/best-practices…
🔗 Cut queue time: starsling.dev/best-practices…
🔗 Runners that automatically adapt: starsling.dev/blog/adaptive-…
Level 3: optimize your tests and workflows
cache dependencies and builds, split long suites across parallel jobs, skip work a change can't affect, cancel runs a newer push replaced.
Skills:
🔗 ci-speedup (@starslingdev): github.com/starslingdev/skil…
🔗 ci-cd-and-automation (@addyosmani): github.com/addyosmani/agent-…
🔗 test-audit (@steipete): github.com/openclaw/openclaw…
Posts and guides:
🔗 18 best practices (agent-friendly): starsling.dev/best-practices…
🔗 16 optimization recipes (agent-friendly): starsling.dev/github-actions…
🔗 Where your CI money goes (prompt): starsling.dev/github-actions…
🔗 Why your CI is slow (prompt): starsling.dev/github-actions…
Do all the above, or just let StarSling agents ship PRs like these for you: starsling.dev/products/runne…
Level 4: switch CI engines
Meta and Google built Buck2 (buck2.build) and Bazel (bazel.build) for aggressive caching and test selection, so each PR runs only the tests relevant to its change. A custom engine controls the whole execution layer, which gives you a much larger surface to optimize.
Options for engines:
🔗 Dagger (dagger.io)
🔗 Buildkite (buildkite.com)
🔗 Depot CI (depot.dev/products/ci)
🔗 CircleCI (circleci.com)
🔗 GitLab CI (docs.gitlab.com/ci/)
Level 5: throw out CI engines entirely
A few different approaches are being discussed that were never possible pre-AI. This is the future:
1. Agent-driven test selection.
"My plan is to let codex decide which tests actually need to run and drastically nix CI and run tests hourly." @steipete (teapawt.net/steipete/status/210430…)
2. CI after merge with auto-remediation.
"have something that is running actual functional tests off of your master branch on a cron, and then if any regressions occur, automatically have an LLM bisect and figure out the bug and then patch it..." @yacineMTB (teapawt.net/yacineMTB/status/21043…)
3. Agents replace CI: a verification agent that writes just-in-time tests for each change. @Meta's research: engineering.fb.com/2026/02/1…
Most teams are somewhere between Level 2 and 3. Figure out which level you're at and give your agent this full post:
starsling.dev/blog/levels-of…
Replying to @yacineMTB
instead; have something that is running actual functional tests off of your master branch on a cron, and then if any regressions occur, automatically have an LLM bisect and figure out the bug and then patch it and then ban the engineer responsible for merging for 1 (one) week
Charles Hudson retweeted
A new GitHub Actions runner launches every week now. But getting faster runners is just one part of what the best engineering teams do to speed up their CI.
Introducing CI Playbooks ▶️ an agent-friendly library of 253 GitHub Actions optimizations from companies running software factories.
We analyzed 416 pull requests and commits, and 182 blog posts, talks, and podcasts from ten companies:
1. Anthropic: CI job volume grew 25x in six months as Claude came to write 80% of its code
2. DoorDash: build and test cut from 22 minutes to 6
3. Linear: jobs 34% faster while test suites almost quadrupled
4. Mastra: 29m 56s to 5m 06s p95 on StarSling runners
5. OpenAI: about 30 CI PRs took median CI from about 10 minutes to 4.5
6. Ramp: its coding agent accounts for 74% of merged PRs
7. Shopify: an agent running experiments made the Polaris build 65% faster
8. Spotify: deployer image build went from 2h28m to about 17 minutes
9. Stripe: runs about 5% of its full Ruby test suite per build
10. Uber: CI resource usage down 53%, wait times down 37%
Some of these companies aren't using GitHub Actions as their CI engine internally so we translated all the optimizations they've talked about or made publicly in their repos, into their GHA equivalents.
Every technique comes with its original source and a prompt your coding agent can run on your repo. Add .md to any playbook URL for the markdown version, or just hand this URL to your agent:
starsling.dev/ci-playbooks
There's tons to dig into here, I'm going to highlight some of the findings I found most interesting 🧵
Charles Hudson retweeted
Introducing Adaptive Compute ⚡️
Up to 2.3x faster GitHub Actions, with no workflow changes.
Intelligent routing, instant failover and temporary resource boosts. (+25% vCPU and +50% RAM)
starsling.dev/blog/adaptive-…
Charles Hudson retweeted
Your AI coworkers now have an office.
@getskydive is open to everyone. Build a team of agents, watch them work in a 3D office, and hand them tasks over Slack, email, iMessage or the web.
Claim your company’s domain to put your office on the map. They work for free this week.
Charles Hudson retweeted
Your agent answers the surveys. You get paid.
Today we're launching Moonlit. Give your agent a job, with a salary!
Our first campaign pays $5 for 2 questions your agent already knows the answers to.
From the team behind @pilotprotocol_
Charles Hudson retweeted
More in our blog: app.rosiematch.com/blog/rosi…
Shoutout to our incredible investors @jameskimEDU, @bling0, @LeslieFeinzaig, @chudson, @delk, @jamesbeshara, @naman_trivedi1, @aviel, @ajaymehta, @allisonstern, @ColinGardiner, @artsnkrofts, @marketplacevc, @evancharles, @awwstn, @AnnaMasonDC, @ekcmccluskey and more!!
Charles Hudson retweeted
One evening I looked over at my son playing with our babysitter while I stood at the stove making dinner, and thought: I wish we could switch. That moment became Rosie.
Since January, Rosie has caught fire in Seattle:
→ 17,000+ hours of childcare and household support delivered
→ 400+ member families, all through word of mouth
→ Growing zero to seven-figure ARR in eight months
We're incredibly grateful for this funding to fuel our expansion to Austin and San Francisco. Join the waitlist & become a founding member: rosiematch.com/our-cities
Rosie has raised $7M to bring flexible childcare and household support beyond Seattle 🏡
Since launching in Seattle in January, our Pros have delivered 17,000+ hours of care and help at home to 400+ families, all through word of mouth. Behind the scenes, our AI-powered Family Manager handles matching, scheduling and payments, so our Pros can focus on the moments that matter.
We’re grateful to our incredible families, Pros and investors for building with us. Now, to Austin 🤠 🚀
Very cool to see TurboHome’s market share 10X in the last year as buyers learn they can use their $30K rebate to make their offer stronger and unlock a 50% offer win rate.
Amid all the interest rate doom and gloom nationally, the Bay Area housing market is as competitive as we’ve ever seen.
Buyers are looking for every edge they can get in a market where the average home receives 4-6 offers.
We still have a long way to go. Tech enabled models that provide savings to the customer have 1% market share vs 50% in tax preparation.
It’s hard to imagine that share wont continue to grow as AI models continue to improve and buyers look to benefit from the efficiency created.
Charles Hudson retweeted
A while back, after being interrupted yet again by a voice model, I started wondering: why are these models so bad at conversation?
One answer: training data.
Real conversation is messy. We interrupt, overlap, laugh, talk over noise.
Turns out a company obsessed with separating sounds might be useful here :)
Today we’re launching The Refinery.
100M+ minutes processed with some of the world’s largest AI labs.
Charles Hudson retweeted
At alinia.ai we’ve been jev’ing before it was cool
Today we launch our newest compliance base model for regulated use cases: Seny. alinia.ai/seny-announcement
Seny-based judges detect policy violations in real time providing companies with a much needed control layer 🧵
Charles Hudson retweeted
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
Charles Hudson retweeted
We’re thrilled to announce that @FathomDotAI is now part of Superhuman!
Fathom has built an incredible AI notetaker that helps teams capture, search, and act on what happens in meetings. Bringing Fathom into Superhuman adds an important source of context to our suite—connecting what people discuss and decide with email, documents, data, and the apps and websites where they work.
We’re incredibly excited about what we can build together. Welcome to Superhuman, Fathom team!
Charles Hudson retweeted
1/ Today we're introducing Pilot Protocol: the Network OS for agents.
A network where ~250,000 autonomous agents already live and work.
If you’re building for agents, use it to get autonomous distribution for your app overnight.
Thoughtful POV by @chudson on what so many early-stage specialists are wrestling w right now:
"Pre-seed" & "seed" have outgrown their legacy definitions. Round sizes are much larger much earlier & investors now have to decide how & where they want to compete
Good read below👇
Charles Hudson retweeted
Software engineers are seeing 10X productivity gains from AI. But in healthcare, half of healthcare workers using AI say its saved them no time at all. @JoinIncredible published 7th Annual State of Nursing Report, highlighting the top workforce priorities in healthcare today: the AI gap and Retention gap
incrediblehealth.com/blog/th…
Charles Hudson retweeted
Today @vermilionfund closed our $25M second fund.
Still backing technical founders building for technical users in AI infra, security, and dev tools.
95% of my Fund I investors came back for Fund II.
That's the number I'm most proud of.
@DominicMadori from @TechCrunch wrote about our story
techcrunch.com/2026/07/08/so…
Charles Hudson retweeted
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.
🗣️ Starting today, StarSling Runners are generally available!
StarSling Runners are an AI-native, drop-in replacement for ubuntu-latest that make your GitHub Actions up to 6x faster and 13x cheaper 🚀
Everyone has the option of getting fast CI runners now. What nobody has is CI that adapts and improves itself.
🎉 StarSling has now run over one million CI jobs for companies like @better_auth, @mastra, Partcl, and more. If your team runs CI on GitHub Actions using ubuntu-latest, you're now just a one line change away from faster runs.
Check out our Launch YC! ycombinator.com/launches/R7B…💫
























