joshribakoff

They spent 2 years building savings after their business failed. Those savings seem to have lasted them around 8 months. And yes there are probably people who work at McDonald’s and can’t afford rent.

From the perspective of the simple dialog, its a problem with the consumer app who calls the api to open the dialogue without context, without informing the user or obtaining user consent prior to doing so. Plenty of apps get this right, eg. Have a screen with buttons in the on-boarding flow that opens the dialogue next to the “maybe later” option, and use that page to explain or “sell” the benefits of enabling the permission.

Maybe the dialogs api should be treated like autoplaying videos with sound on the web, just disable the api if it is called outside of a click event handler, and let the consumer app break. That would be a breaking change for OSs tho. It wouldn’t prevent malicious UI patterns from triggering annoying permission prompts, but it would minimize accidental bad ux eg calling it in the main method on load

Instead of a simulator that runs in its own process, I would recommend mocking out inline in the tests. Only then are your tests entirely “self contained”.

Otherwise your test implicitly make assumptions like “User A is always off-line” (or bucket A is always empty), add a reader of the test would have to go find where that assumption is hardcoded in the simulator and verify it separately from the test itself.

To someone reading that test it is not intuitive that user A is always off-line or bucket A is empty.

If you mock out in line, you can have tests that mock user a / bucket a — to simulate any state for any user or bucket. Perhaps multiple states in one test.

Also think about scenarios like the bucket is initially populated, but then on a subsequent request, it’s empty.

With the simulator approach, you’d have an endless number of permutations of scenarios to add — each of these scenarios would be highly coupled to the test it belongs to, so why not just put it in line in the test itself?

Any reader of the test that uses inline mocking would clearly see which scenario is being tested instead of having to dive into a separate simulator to see what special treatment is hardcoded for User a / bucket a (or maintaining potentially hundreds or thousands of simulated buckets with ad hoc names like “empty-on-second-request” or “bucket–that-responds-slow-and-times-out-but-succeeds-on-retry”.

You seem to be mixing up integration tests with end to end tests.

It’s more so the divide and conquer. Recursion vs iterative is tangential.

He didn’t post in the thread or email any candidate. He copied a public comment from one public site to another.

When he goes on TV and claims the models may be conscious or may automate 90% of SWE in 6 months it is designed to pump the stock price. Its ironic they hold a fictitious scenario over candidates heads like “would you still keep working here if our revenue were $0 and we cant make payroll” (the realistic scenario where the stock price literally drops to $0)

Oh please, you sound so disingenuous, the claim wasn’t that the documents contained a specific phrase. You’re moving the goal posts. Their name is literally a play on anthropomorphising the models, such as… the ceo going on tv shows and repeatedly saying the models may be conscious and they may start nuclear wars, etc.

You’re right — people really should think critically, but the issue remains, that many do not.

That’s standard click tracking usually. No different than clicking on a sponsored Google result.

I just prompt my ai coding assistant directly to scaffold out problems and test cases.

One nice thing is the LLM can “reason” about my weak areas and tailor the drills it generates next to keep me on a productive path.

I personally don’t get much value from pulling in 1,400 arbitrary problems but this is cool for people who want to hit the ground running.

Part of the entire premise of the article is that many people aren’t being given a choice, so this is a really weird deflection. I disagree about the author not recognizing a middle ground.

An LLM is not what the public was “sold”, they were “sold” a personified agent. That is the gap.

Some people have the agent hallucinating its own meaningless tests

Isn’t there a gray area between unregulated markets and regulated markets?

A corrupt regulator could prohibit this for ride share industry only, to target under a pretext.

Manipulating market incentives could also reduce competition assuming an end goal is acquisition.

Of course with no regulation, corrupt players can consolidate power.

I would say unchecked power is more dangerous, so the question to me is if we presume regulation is good then how do we structure it so the regulators are regulated?