I'm on SuperGrok ($30/month) because that gives you access to Build. Unless they changed it, you don't get access to the CLI tool with the basic subscription.
Most of our group was slow to start using coding agents. That is over now. Almost everyone is using them, and stories are completing a lot faster -- and our testers are getting crushed. They still do most of the work by hand, so they are really falling behind.
I keep wondering what this looks...
Thought this was pretty okay, especially considering the runtime. If it had dragged on for two-plus hours, forget it. As a short and sweet action/sci-fi movie, though, it works fine.
Of course they can't help but to sprinkle in the gay shit.
Again, I don’t understand why some of you think Reacher is an ultra-realistic TV show. I enjoyed this season for its over the top action sequences. That jump-on-the-table move was awesome.
I agree with them. I saw this in the theater when it was released. It got a few laughs, but I remember something feeling off even then. Their discussion put a name to those problems.
I played this years ago and thought it looked incredible. There just was not much game at the time.
I also remember it as a hosted session more than a persistent world. Is that still how it works after 1.0, or are people running dedicated servers now so the world stays up when nobody is logged...
For us, we just got a token bump to "60,000" credits per month -- whatever that equates to. All I know is, we were running half that up until then and it was apparently costing the org couple hundred bucks a month. I mean, that's pretty cheap all things considered. Our new junior dev is...
Yeah, but then it would be no different than a single tile on a virtual desktop. But you're right, there will be a delay when switching modes. Not sure how annoying that will be. Regardless, fun project so we'll see how it goes.
I don’t know... that’s actually a pretty cool idea. It is a way of rebooting the franchise without quite rebooting it. Older fans would recognize the characters and actors, while it would still feel reasonably fresh to newer viewers who never watched the original series. They could return to the...
I have only used tiling window managers a little. But I'm considering creating a compositor built around screens rather than tiles.
Launch an application and it gets an "Application Screen": exclusive use of the monitor. No tiles, no floating windows, no decorations. Save dialogs, modals, extra...
Bernard suggested that’s not a machine (AI) but an actual person they’ve been talking to.
I’m also wondering, given the "memory-wiping" vitamin D, whether they haven’t been in the silos as long as they think.
Highly recommend this mod: https://www.nexusmods.com/baldursgate3/mods/14430
This mod has breathed new life into the game for me. Even on Honor Mode, the base game is a cakewalk now. This one adds genuine difficulty and situations where you really have to use every tool in the toolbox. I had...
Yeah this sounds super interesting. Just don't have a system to try it out on. Hate to screw up my current CachyOS set up which has been working awesome.
I love this series, although, like everyone has said above, seasons 1 and 2 were absolutely the best. I didn’t bother with season 4 with Chris Rock… it seemed like it was being panned at the time and was during the height of the woke shit.
I also use: GitHub - oraios/serena: A powerful MCP toolkit for coding, providing semantic retrieval and editing capabilities - the IDE for your agent
Which apparently improves efficiency.
I had Opus 5 code review Grok 4.5, Grok read the code review, do the changes, back to Opus review the changes, new code review, and then update. They went around in circles and burned nearly my entire week token allotment.
A few weeks ago, I had GPT-5 generate a mockup of a healthcare application dashboard. The result blows away our enterprise app and is, quite frankly, embarrassing.
I have the agent look over a PR and write its findings to a separate markdown file that I review myself. It can get pretty nitpicky, so I just pull out the things that are legit concerns.
One thing I've done for the junior dev is add special instructions in the "investigation" workflow component. The agent looks at git commits, pulls story details from ADO, and so on, so it can gather more context on why things were implemented a certain way or why changes happened. The...
I've taken a junior dev under my wing and have her using my custom Copilot workflow. The process I built is pull story -> investigate -> plan -> implement, with each step gated by human review. I keep stressing that she needs to actually read and understand what the agent writes (it dumps each...
Truthfully, we are still in the infancy of all of this. What the landscape will look like in another year or two will almost certainly blow our minds.
“This is the worst it will ever be.”