Hacker News Reader

Wrong, actual code and tests provided than show 10000x speed up. Users can run it themselves and have been seeing better results.

Appreciate if you didn’t make up stuff.

View on HN

As I pointed out the 10000x speed up claim is smoke and mirrors, and any of your team could have spent 10 minutes and figured that out by profiling the code. It's a silly claim that doesn't hold up to scrutiny and detracts from your project by setting off the bullshit detector that most programmers have on marketing hyperbole. It's not the flex you think it is.

You may not have made up the number, but you did invent a contextual framing where you could claim that number accurately. But the framing is not a useful or practical one. It's like saying that your car is faster because the driver can turn the key in the ignition more quickly.

Instantiating an agent is not the bottleneck for LLMs. Two hundredths of a second is a rounding error compared to what the model costs in time.

Reply on news.ycombinator.com