Danver Braganza

0

7

126

Danver Braganza@404Cause·14h

@lauren__dash @MRTIBBETS And on his first day in the office, wasn't it?!

English

0

6

Lauren Dash@lauren__dash·15h

@MRTIBBETS @404Cause Great job naming it @MRTIBBETS 🤠

English

0

1

8

Alexander Tibbets@MRTIBBETS·21h

Really great work @404Cause. Happy to see Offload in the world making a difference, one burst of test runs at a time. 💥

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

0

5

134

Danver Braganza@404Cause·17h

@thesquashSH @bernhardsson I'm ambivalent, TBH. But although we built Offload for the agentic loop, it's been difficult to resist the urge to use it for CI as well.

English

2

11

Nick Sweeting@thesquashSH·21h

@bernhardsson Honestly I really hope Github Actions just gets better, I want Github to win so bad because I hate having to use 8 different tools but the reliability is just so ass right now

English

2

0

1

152

Erik Bernhardsson@bernhardsson·21h

I love this. Tests are a class of “embarrassingly parallel” computer problem and scaling out makes so much sense. Next step: GitHub Actions replacement

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

13

6

254

37.7K

Danver Braganza@404Cause·17h

@charliermarsh Congratulations, @charliermarsh and team!

English

5

Charlie Marsh@charliermarsh·1d

We've entered into an agreement to join OpenAI as part of the Codex team. I'm incredibly proud of the work we've done so far, incredibly grateful to everyone that's supported us, and incredibly excited to keep building tools that make programming feel different.

English

284

145

3.1K

434K

Danver Braganza@404Cause·17h

@MingtaKaivo @bernhardsson The key pain point we built Offload for was verification in the agentic loop. But we've been feeling compelled to use it for CI as well.

English

0

1

11

Mingta Kaivo 明塔开沃@MingtaKaivo·21h

@bernhardsson GitHub Actions is the obvious next target. the pain isn't the test execution — it's waiting 8+ minutes for a CI queue before you see if the build even passes. distributed test sharding closes that without changing anything about how you write tests.

English

0

2

220

Danver Braganza@404Cause·17h

@LeoTava8 @bernhardsson Agreed, and that's why our CLI and bundled skill enable agents to retrieve specific failure traces. #debugging-failed-tests" target="_blank" rel="nofollow noopener">github.com/imbue-ai/offlo…

English

1

8

Leo Tavares@LeoTava8·19h

@bernhardsson we've spent a decade debugging yaml. agents actually reasoning about failure logs is the upgrade we need.

English

0

2

132

Danver Braganza@404Cause·18h

@ctatedev How are your agents running your tests? We created Offload to get your tests off your machine into the cloud, but there's no reason why you can't fan out starting from a sandbox. imbue.com/product/offloa…

English

7

Chris Tate@ctatedev·1d

~100% of my dev is done in sandboxes in the cloud Highly recommend it: - Unlimited parallel agent sessions - My local machine stays safe - Can work from anywhere - Can close laptop - Lap stays cool Interesting idea to visualize with Kanban

Ryan Carson@ryancarson

100% of dev is going to be done in sandboxes in the cloud, controlled by kanban boards. Trust me, I love my local machine and gorgeous mac apps, but all of it is just a terrible form factor for running a team of agents effectively.

English

63

37

945

157.5K

Danver Braganza@404Cause·18h

@TomasPiaggio @modal @imbue_ai Only 6x? Why limit yourself?

English

2

17

Tom Piaggio@TomasPiaggio·21h

@modal @imbue_ai now i can hit my rate limits 6x faster.

English

2

0

2

88

Modal@modal·21h

With Offload, @imbue_ai saw a 6x drop in execution times by giving agents the ability to parallelize tasks. Now open source, and backed by Modal Sandboxes 👇

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

4

43

5.4K

Danver Braganza@404Cause·18h

@podviaznikov Hard to disagree that this is where things are heading.

English

3

anton 🇺🇦🇺🇦🇺🇦@podviaznikov·19h

that is exactly how everything will be done. parallelized compute. for CICD. for agents. whatever

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

2

0

4

270

Danver Braganza@404Cause·19h

@FileCityAI @imbue_ai @modal Hey @FileCityAI, I'm one of the developers of Offload here. Nice work, thanks for providing this visualization and walkthrough! Have you thought about visualizing the dependencies between the Rust types as well?

English

0

2

45

FileCity@FileCityAI·19h

@imbue_ai @modal Made a File City Tour of Offload A visual map + narrated walkthrough of the open-source Rust CLI that spreads test suites across 200+ @Modal sandboxes for ultra-fast parallel testing github.com/imbue-ai/offlo… Great work! @imbue_ai

English

0

8

519

Imbue@imbue_ai·23h

Your parallel agents needed scalable test coverage yesterday Introducing Offload: a Rust CLI that spreads your test suite across 200+ @Modal sandboxes, freeing your CPU to keep your agents shipping. On our Playwright suite, it took a 12 min run to 2, at $0.08 a run

English

14

16

176

65.7K

Danver Braganza@404Cause·19h

@olvrgln @imbue_ai @modal We have an internal workflow with two image ids. A base image which updates only when your deps change, and a current image which is updated with local source changes. This saves on Modal Image rebuilds. On our roadmap to release in a user-friendly way.

English

2

24

Danver Braganza@404Cause·19h

@olvrgln @imbue_ai @modal Yep, the sandbox image contains a copy of your repo + installed dependencies + local changes. That way each of these sandboxes doesn't need to waste time cloning the repo from your laptop or source control.

English

3

0

2

66

Danver Braganza@404Cause·19h

@theplgeek Thank you, @theplgeek! I'm eager to hear how it works out for you, and if you have any feedback.

English