NorthSouthNorth

They apparently rolled this decision back.

The thing is, once you start paying for it, then a service like blacksmith is so attractive and easy to replace gh, that it seems a no brainer. We swapped like 30-40 gh actions from gh runners to Blacksmith runners and cut our ci times by about 4x on almost all of our workflows. And all you do is auth their gh app and update an image tag.

Especially if you can take advantage of the high vcpu runners, then you can get some really fast pipelines going (not affiliated w/ them, just an enjoyer).

I remember a year ago and a bit ago trying to get an api key for some Gemini ~3 model. I probably spent an hour trying to find how to get a key in AI Studio. Finally figure it out and got a key.

I needed batch mode. Oh but the key from service X doesn't support batch mode, you need to go service Y and btw there is no UI for that so you need to use an SDK or API (cant even remember anymore). Spend an hour figuring that out, then it turns out I can't use batch mode because I don't have permissions. Have boss spend an hour granting me permissions. Then service is unavailable every n requests for 5 minutes, so I have fuck with exponential back offs.

Glad to see they moved that experience to their consumer offerings lol.

Not to be confused with ACP (Agent Client Protocol).

Got the m5 pro like last month. Actually no FOMO this time. I wonder how the AMOLED looks irl, I bet it's sweet but the current pro motion is already top notch.

Title reads as if they just cut down the entire thing lol. (The original title too).

My 16" M1 Pro personal laptop is literally just as smooth for 99% of tasks as my work M5 pro. It's crazy how well they built these. I mean the screen on this and the speakers are superb. And you can find these for like 500-800 bucks 2nd hand. I just hope they give us a few more years of MacOS updates.

Looks like a cool monitor (900 euros here). I am happy to see that there are more and more of products in this category popping up. I wish they would chose better names though (Kuyon is pushing it, but Japannext is very Aliexpress coded).

Man, but these Macbook Pro screens are so sweet.

Everyone is entitled to their opinions I guess :P

I never cared about any of this until I had to help my wife with monitor shopping where colour accuracy & PPI is king, and spent some time testing these out. Hard to go back now.

I was using the 49" Samsung Ultrawide for years at work and now I feel like I can count the pixels.

One almost forgets how easy it is to read a wall of text when it was written by a human. This was nice to read (also because you can feel it's got care behind it).

I kinda agree with you but this is for people who don't want to think about it. That 24" monitor is incredible compared to anything sub 500.

A 32" gaming monitor is garbage once you're used to 216 ppi for creative work.

Realistically there is like 1 display from BenQ that even comes close spec wise.

Their target user will pay 1700 and have an incredible machine that does literally everything they want and probably won't need to upgrade for what, 5-10 years in a single purchase. Compared to having to buy a mouse, keyboard, speakers, webcam, and monitor.

Also having good speakers and a webcam is underrated. I use my wife's Studio Display sometimes and it's a real, real nice piece of hardware. I have a 5k LG ultrafine at work that's spec for spec similar but it feels like a toy in comparison (it's also like half the price though so).

Yeah like the other say it depends on the harness. "into chatgpt" could mean many things. It's likely not predicting the next tokens to emit but rather thinking "okay the data is in format x i need to map it to y, and also this is a good fit for statistical model z" and then writing and executing a python script that runs whatever statistical model on the inputs and then produces some other output file that it then uses to produce a report.

That being said, the correct way would be to just use the LLM to produce the script, validate it themselves (often times there's something very minor wrong), and then use that script next time instead of an LLM to rewrite the same script over and over every time you need to produce a report.

Yeah - the only time it's worth it is if you're also learning. I volunteer extra time when a project really teaches me something new (or there's some fire and we need all hands, but that's rare).

This year I had the chance to learn IaC and generally using AWS through a project, but I had to donate a bunch of my time learning. Now that's 2 things I can claim I have worked with in production. But fixing bugs and general dev work, no thanks.

It's hard to read old books sometimes. Unless you grew up doing it you just simply don't understand character intentions. Reading a Dostoyevsky book it's hard to relate to characters or even understand the outrage in a lot of scenes.

If you don't know your history well, the whole setting for the Count of Monte Cristo is confusing af. So you now need to read up on the fall and rise of the monarchy and Napoleon, and his exile etc. Also a lot of words are just foreign to most people.

Where as reading a book written recently it's just way easier to get into it since it's written in the context of our collective memory (if that makes sense).