eddyzh

Thank you! It's cool indeed.

This resonates a lot with me, well worth the long read. And the last part about how to try keep the human part relevant, is exactly what I try for now.

This checks out 50-250kg CO2 are common guesses for 10k spend.

That would be in the same order of magnitude of an average daily commute with an older car.

The fact that is is ai generated.

For that test you have to compare letting a fresh agent (subagent) or the same model do the same review.

The fact that a review helps does not prove the model choice for the review.

You reviewing your own writing helps too!

Added the () part since this title has been used a lot in other articles.

Exactly this.

This may be worth the discount. Or not if your time and attention is worth (quite) a lot.

At work I use opus max Fast It hardy ever fails for no reason even if I forget to give it all the right context. At home i run sonnet, and it does not get what I meant or expected 20-35% of the time. Due to the enormous difference in cost, depending on the value of your time (hourly rate) that might be a nett benefit.

Sonnet being faster alone would not be worth the failure rate for me.

At home i just not want to pay more than 20 bucks for incidental projects.

And opus max would just consume my tokens in one round.

Thank you. This is great to hear!

While chatGPT was not out then, the ML that drives robotics was acting by then very much.

In had one app like that from Cydia Loved it.

Would that be unified memory? Where the gpu and cpu can share the memory? Which is key for performance.