lemonlearnings

Lots of fish? Send to microservices teams. Few fish? Monolith team.

I noticed all the fish. Not because I'm holistic though. I first noticed the big ones. Then though "hmm what's the trick" and carefully scanned the rest of the image because of that. But I guess the experiment was done in a more controlled way. Without "fish test" in the title!

Isn't grep + LLM a form of RAG anyway?

So Google, Amazon, Oracle, Microsoft etc.

Half true, Half snuffing out lamps to trick you.

Seems like a ME/CFS simulation too. Well done.

Adding the compute story would be interesting as a follow up.

Where is that done? How many GPUs do you need to crunching all that data. Etc.

Very interesting and refreshing read though. Feels like what Silicon Valley is more about than just the usual: tf apply then smile and dial.

.xlsx is the key. Can use it with sheets, M$ or libre office. Or unzip it and take a look at XML.

Google takeout is key. Why struggle backing up individual Google things.

Mattresses tend to be better at home than in prisons

Nice. Find accomplices by leveraging the power gradient of employment. Giving them legal advice to boot. Think that should add more time. It's an evil thing to do.

Hence paragraph 2

Should it be a client error "vote differently" or a server error "this server is not licking the boot" ;)

So the entire internet goes off during La Liga?

Ethically dubious. Paying someone to poison themselves to test your site. That they are up for it aint an excuse.

If it was "ill test your site when I next come home from a big night out" it may be OK.

There is a lot more money to be made if you can attract a cohort of suckers. It is basically an entirely different business model. Drug dealer rather than an eBay. And running a betfair cannibalises the sucker arm.

One says

> People at highest risk are those who work in environments where they’re continuously exposed to fragrances, such the cleaning industry, cosmetics industry or agriculture industry. You also may be at slightly higher risk if you are continuously exposed to fragrance through personal overuse.

And do the thing at dawn if you can!

Or get a dog.

Australia Kmart got a good round of applause recently on HN and I think it was because of this reason. MZOO prices but with Kmart brand curation (they use their own brand name Anko alot but under the hood must be different suppliers etc.). Not sure how they do it and pay all that staff and rent, but they do!

Also the maga tariff and bizarre interventions age.

Ah the hitman business model

The car analogy is Tesla needs to make tractors to compete.

I think there is a bitter lesson to the bitter lesson.

Sure you can throw more compute at it. But it cost a lot of money and you hit resource limits.

We have been doing an end run around the bitter lesson with prompt engineering. Also by using different models for vision vs. text. By getting (human coding) agents to "think" and run code.

The bitter lesson might be that you cant predict what the thing is that will be most optimal tomorrow and any player in the AI game can be innovated out of existence at any time.

Maybe anyone except TMSC.