Founder of @currents_dev - performance, monitoring and analytics platform for Playwright. Creator of @sorrycypress.

Vancouver, BC
Joined October 2010
"slopycat" - a poorly made AI copycat
1
78
Hope to see some positive changes in BC! @Conservative_BC all eyes on you.
30
Everyone should read OpenAI’s report and be blown away. openai.com/index/securing-ou…
3
161
Same for us - we use @currents_dev to share evidence videos, playwright traces and screenshots as evidence on PRs, Linear tickets and as a demo / preview on Slack. Your tests are the evidence.
We now require all PRs to include a screen recording artifact. It's the only way to verify that an agent's change was tested end-to-end. You shouldn't have to check out a change locally to trust hitting merge.
1
1
183
Flying fucking cars, really!?!
1
96
Verification engineering and CI bottleneck is a hot topic right now. Existing paradigms shifted: - diffs matter less, outcomes matter more - reviewing means examining outcomes, not diffs - rich artifacts like rich traces, network logs capture in isolated environment enable effective iteration 1/5
With 451,000+ downloads, Currents touched its highest ever weekly downloads last week. We realise that some of you aren't aware of what we've shipped, that got us here. So, here are 10 features that our engineering team has shipped in the recent past: 1. Shareable failure context Every failure now gets one permanent link that holds everything: the error, the trace, the logs, and the branch and pull request it came from. Open that link, or hand that context to an AI tool in one click, from the dashboard, the IDE or Slack, and have your AI fix it! 2. Evidence sharing for agents Currents' MCP server lets your coding agent attach a trace or a whole run to a PR or a ticket. A reviewer opens a hosted trace player and sees the proof, instead of 'trusting' the agent. 3. Share a whole run A publicly shareable link that carries a whole run with every test in it, with an expiry duration of your choice. Each test also gets its own public context link. 4. Remote MCP server We host our own MCP server, so Claude, Cursor and Codex connect without a local install. You sign in with OAuth, choose the permissions on the consent screen, and get going in no time! 5. One-click IDE sign-in Our IDE extension lets you sign in through the browser in one click. No longer pasting an API key into settings. Click one button and start shipping! 6. Filter by environment Runs, Insights and Explorers all carry an environment value, and you can filter each of them by it. A team that runs the same suite against staging and production can now read the two sets of results apart. 7. New Run summary view The run page is now a flat test list with grouping, sorting, and filters for status, tags, team and project. A details sidebar opens the trace or starts a fix without leaving the list, and your grouping and density choices are remembered. 8. Richer Slack notifications Run notifications now use colored attachments, update in one thread instead of many messages, and post a recovery message after a failure clears. Each failure message carries a Fix with AI button, and its link opens the run summary with the matching filter already applied. AI all the way! 🤖 9. Dashboard refresh Settings, Reports, Test Suite, Code Coverage, Billing and Usage, API keys and Manage Team all moved to a shiny new interface, and is getting lots of love! 💅 10. Set artifact retention An organization can set its own artifact retention policy, for a whole run or per-artifact type. Teams with large traces control their own storage instead of accepting one fixed window. And these are just the highlights. There are many, many, many more changes to make your or your agent's Currents experience delightful like: - Scoped OAuth API access - Rich Linear integration - Richer runs for our REST API - Jira linking over the API - AI Test explanation in the Currents dashboard sidebar - Github run comments - Email domain verification (for member management in organization) - AI usage consent controls For admins, we have: - Self-serve SSO setup with a wizard, - Okta catalog app, - SCIM token rotation, - Email domain verification (work domains), - AI consent setting And so much more! In fact, we're cooking some awesome features right now, that are gonna release soon! Big things incoming! Till then Run your test suites and ship with CONFIDENCE! 🌊🚀
1
1
2
137
@currents_dev customers have been already iterating efficiently: agents build and use rich artifacts to verify and iterate on either fixes or new features. - Record rich artifacts from your testing suite - Provides agent-ready API to fetch and explore recordings docs.currents.dev/ai/agentic… 4/5
1
15
The recent release of Currents Sessions closes the gap for ad-hoc recordings. Instead of creating a new tests, just throw your agent's recordings to Currents and make it available everywhere: > currents session start > currents session attach ... > currents session attach .... > currents session share docs.currents.dev/ai/agent-e… docs.currents.dev/ai/referen… 📸 🎥 💿 🫆 5/5
1
9
Andrew Goldis retweeted
With 451,000+ downloads, Currents touched its highest ever weekly downloads last week. We realise that some of you aren't aware of what we've shipped, that got us here. So, here are 10 features that our engineering team has shipped in the recent past: 1. Shareable failure context Every failure now gets one permanent link that holds everything: the error, the trace, the logs, and the branch and pull request it came from. Open that link, or hand that context to an AI tool in one click, from the dashboard, the IDE or Slack, and have your AI fix it! 2. Evidence sharing for agents Currents' MCP server lets your coding agent attach a trace or a whole run to a PR or a ticket. A reviewer opens a hosted trace player and sees the proof, instead of 'trusting' the agent. 3. Share a whole run A publicly shareable link that carries a whole run with every test in it, with an expiry duration of your choice. Each test also gets its own public context link. 4. Remote MCP server We host our own MCP server, so Claude, Cursor and Codex connect without a local install. You sign in with OAuth, choose the permissions on the consent screen, and get going in no time! 5. One-click IDE sign-in Our IDE extension lets you sign in through the browser in one click. No longer pasting an API key into settings. Click one button and start shipping! 6. Filter by environment Runs, Insights and Explorers all carry an environment value, and you can filter each of them by it. A team that runs the same suite against staging and production can now read the two sets of results apart. 7. New Run summary view The run page is now a flat test list with grouping, sorting, and filters for status, tags, team and project. A details sidebar opens the trace or starts a fix without leaving the list, and your grouping and density choices are remembered. 8. Richer Slack notifications Run notifications now use colored attachments, update in one thread instead of many messages, and post a recovery message after a failure clears. Each failure message carries a Fix with AI button, and its link opens the run summary with the matching filter already applied. AI all the way! 🤖 9. Dashboard refresh Settings, Reports, Test Suite, Code Coverage, Billing and Usage, API keys and Manage Team all moved to a shiny new interface, and is getting lots of love! 💅 10. Set artifact retention An organization can set its own artifact retention policy, for a whole run or per-artifact type. Teams with large traces control their own storage instead of accepting one fixed window. And these are just the highlights. There are many, many, many more changes to make your or your agent's Currents experience delightful like: - Scoped OAuth API access - Rich Linear integration - Richer runs for our REST API - Jira linking over the API - AI Test explanation in the Currents dashboard sidebar - Github run comments - Email domain verification (for member management in organization) - AI usage consent controls For admins, we have: - Self-serve SSO setup with a wizard, - Okta catalog app, - SCIM token rotation, - Email domain verification (work domains), - AI consent setting And so much more! In fact, we're cooking some awesome features right now, that are gonna release soon! Big things incoming! Till then Run your test suites and ship with CONFIDENCE! 🌊🚀
3
2
3
4
1,071
A mandatory drill for Business Continuity plan
37
Meanwhile in Canada…. 🦗
We announced an agreement to acquire a nationwide low-band spectrum license portfolio that will pave the way for @Starlink to become a major mobile carrier in the US. With this new spectrum and our Gen2 constellation, Starlink Mobile can ensure Americans have access to high-speed mobile broadband no matter where they are → spacex.com/updates#starlink-…
34
Life update: Excited to share that I've joined @currents_dev ✌️ AI has made shipping easier than ever. Shipping reliably, however, is still a challenge. Our goal is simple: Help devs deliver software reliably. Big things coming! 🌊🚀
30
4
62
2,018
Not only they are raising taxes to cover overspending, but also causing this tz mess. 🤬 I am voting for whoever cancels this BS #bcpoli
37
Andrew Goldis retweeted
Nobody wants to use your agent. Give the tools to my agent to handle it e2e
Everyone got a coding agent. Nobody got a QA engineer. Until today. Meet Ship, your Autonomous Quality Engineer. It tests deployed PRs, reproduces bugs from Slack and Linear, and hands Claude or Codex the context to fix them. Try it free: ship.contextqa.com
2
3
24
3,414
Currents.dev is the missing half of “give the agent a way to check its work.” It turns E2E runs into structured evidence—traces, failure snapshots, history, flakiness—then exposes that via MCP so the agent can triage new regressions vs known flakes, fix, and re-check the next run.
To make your agent produce high-quality output, equip it with a way to check its work. Tests are part of that. Here are some of the ones I think are worth investing in. - Write end-to-end tests that simulate real user flows and act as the ground truth - Use property-based testing to specify what should never happen and generate thousands of test cases that attempt to break those constraints - If replacing a system, compare the old and new versions using randomly selected inputs - Require tests to run quickly and deterministically, which reinforces the agent to retry when the loop is slow or unreliable (the agent may learn to retry, not fix the problem) Example tests say what should happen; properties say what must not.
31
Andrew Goldis retweeted
Jev will probably go down in history as the most thoroughly spanked startup ever — a brutal reminder of how few ways are are to create a competitive moat in software these days. (Hint: if you don’t have proprietary data that clearly distinguishes your offering then watch out.) 1. Launch a service that popularises a novel and valuable way to use LLMs 2. Watch literally everyone clone it almost instantly, in many cases beating it
Cloudflare's Clef beat Jev in just two weeks How: "frozen" Qwen does one prefill pass, a tiny schema head scores every answer option in parallel, with no text generated at all. That's why it's 4x faster than Jev at 2x the accuracy on some benchmarks. And it's open source! Link to HF: huggingface.co/Cloudflare/cl…
167
185
55
3,283
471,531