tomatohs

6 ms·

1.
by tomatohs 1mo ago
Decline? AI coding agents drove commits up 14× in a year. Pull requests opened by AI coding agents surged from roughly 4 million in September 2025 to more than 17 million in March 2026 — a 325% increase in six months. [1] [1] https:/&
2.
by tomatohs 1y ago
TestDriver.ai | https://testdriver.ai | QA, Engineers, Customer Support | Austin / Remote | Full-time / Part Time 95% of companies are still wasting time manually testing due to shortcomings in Playwright, Cypress, and
3.
by tomatohs 1y ago
Computer-use agent for testing: https://testdriver.ai
4.
by tomatohs 1y ago
I started a company to do exactly this a few years ago, and got to work with amazing companies testing their developer experience. The problem is not the docs, it's Conway's law. One team designs the API, the other team designs th
5.
by tomatohs 1y ago
It's not 2010 anymore. Most startups can't even attract "the best engineers" much less hire them. This is the late game, why would an engineer work for a fraction of a percent of equity and a below market salary when the
6.
by tomatohs 1y ago
TestDriver.ai | https://testdriver.ai | QA, Engineers, Customer Support | Austin / Remote | Full-time / Part Time 95% of companies are still wasting time manually testing due to shortcomings in Playwright, Cypress, and
7.
by tomatohs 1y ago
Computer-Use Agent for QA Testing https://testdriver.ai
8.
by tomatohs 1y ago
This is exactly our workflow, though we defined our own YAML spec [1] for reasons mentioned in previous comments. We have multiple fallbacks to prevent flakes; The "cheap" command, a description of the intended step, and the origi
9.
by tomatohs 1y ago
Would love to use this for TestDriver, but needs to support Windows :*(
10.
by tomatohs 2y ago
TestDriver.ai | https://testdriver.ai | QA, Engineers, Customer Support | Austin / Remote | Full-time / Part Time 95% of companies are still wasting time manually testing due to shortcomings in Playwright, Cypress, and
11.
by tomatohs 2y ago
I agree with you IRT scale but not speed. Copy and demo videos are essentially one way communication channels ("fire and forget"). The creator has no idea if the message was understood. Also, writing copy or making a video typical
12.
by tomatohs 2y ago
A friend described calls as "high bandwidth information transfer." An average typing speed is 40wpm but an average conversation is between 120 - 150 wpm so about 3 - 4x bandwidth. Calls also offer sub second latency and maximum pr
13.
by tomatohs 2y ago
Cool! Before building a full test platform for testdriver.ai we made a similar sdk called Goodlooks. It didn't get much traction, but will leave it here for those interested: https://github.com/testdriverai/goodloo
14.
by tomatohs 2y ago
TestDriver.ai | https://testdriver.ai | QA Engineers, Sales | Austin / Remote | Full-time / Part Time 95% of companies are still wasting time manually testing due to shortcomings in Playwright, Cypress, and other frame
15.
Show HN: GitHub bot that plays with your app on a VM(github.com)
1 points by tomatohs 2y ago|1 comments
16.
by tomatohs 2y ago
our agent is available via NPM: http://testdriver.ai
17.
by tomatohs 2y ago
> It is very helpful to give it things like: - A list of applications that are open - Which application has active focus - What is focused inside the application - Function calls to specifically navigate those applications, as many as po
18.
Show HN: TestDriver, cross platform CLI to let OpenAI automate your desktop(github.com)
9 points by tomatohs 2y ago|1 comments
19.
by tomatohs 2y ago
> unshakable foundations in reality Yes, you are correct that it entirely lays in the reputation of the AI. This discussion leads to interesting question, which is "what is quality? " Quality is determined by perception. If we
20.
by tomatohs 2y ago
This comes up all the time. It seems like it would be possible, but imagine the case where you want to verify that a menu shows on hover. Was the hover on the menu intentional? Another example, imagine an error box shows up. Was that correc
21.
by tomatohs 2y ago
We do AI E2E desktop, sent you an email.
22.
by tomatohs 2y ago
It's absolutely not the bottleneck. OpenAI can process a full resolution screenshot in about 4 seconds.
23.
by tomatohs 2y ago
I predict an all out war over deterministic vs non-deterministic testing, or at least a new buzzword for fuzzy testing. Product people understand that a cookie banner "shouldn't" prevent the test from passing, but an engineer
24.
by tomatohs 2y ago
Curious what happened to the other YC Mobile AI E2E company, CamelQA (YC W24). They pivoted to AI assistants. Could be good lessons there if you're not already in touch with them.
25.
by tomatohs 2y ago
You're totally right here, but "debugging failed tests" is a mature problem that assumes you have working tests and people to write them. Most companies don't have the resources to dedicate full engineer time to QA, and
26.
by tomatohs 2y ago
This is the right idea and how we do it at TestDriver.ai. The deterministic selector still has about 20% fuzz matching rate, and if it fails it trys to recover.
27.
by tomatohs 2y ago
Oh no, it doesn’t even run on your local machine if you don’t want it. It runs on VMs.
28.
by tomatohs 2y ago
An AI that can control the desktop. It can see the screen and use the mouse and keyboard. We’re selling it to developer teams to help with QA http://testdriver.ai
29.
by tomatohs 2y ago
TestDriver.ai ( https://testdriver.ai ) | Growth | Full-time | Austin or Remote TestDriver helps engineering teams move quicker by offloading QA to an AI with their own development environment. TestDriver is not just generative te
30.
by tomatohs 2y ago
AI makes things up so it's not appropriate for deterministic solutions. I'm sure something is coming for your mechanical linkage problem. As for using it more, think of it as a co-worker and not a chat bot. There was a switch that

More ›