It’s easy to lose track of time with these tools, he told Business Insider. Soon enough, the girls’ practice has ended, and the parents flood into the changing room. He joins them — with his laptop ajar, so that the AI agent can keep running.
“I have to put it up on a shelf,” he said. “I’m untying my girls’ skates while looking back like: Is it done?”
I just feel bad for these poor kids. You can’t leave your goddamn Claude at home while you take your kids to the rink?
Ohoho, I know exactly how to burn a silly amount of tokens if I want to, which is why that metric is absolutely garbage - arguably worse than ranking developer performance by SLOC committed.
Have it write all of your logging code for you, it may be inaccurate but it is the least damaging place to take the hit as you can just manually search in the source code for where the print was from. They always do something stupid and non uniform making most statements traceable indirectly.
If you fully inform them on the situation, they will be cool with it. We are all in this shit together, well until everything falls apart. Add a [<linenumber>] to the prefix on each statement, who the fuck cares. It was the LLM that did it, not you and you were forced to use it.
I mean this is why I don’t mind using AI, it gives me the time to focus on the people around me.
Also I see it as a sign of a lack of support and isolation in the workplace. Like “I need help, I don’t have anyone to turn to cause everyone is fired so this is my only support” and the anxiety that brings.
I’ve been suffering from too short an attention span. I keep ending up. scrolling on my phone while waiting for ai to spew its slop, and have been caught too many times
Yeah I just swap to the next project. And if all of those are hydrated I go to my hobby projects, and yeah those I don’t use agents for so I just stay there for a while after that lmao
or you can literally just do what is it? something+tab? and it just goes to town without the need for you to confirm anything.
I mean it’s gonna turn out slop that won’t scale and be full of exploits anyways regardless if you auto confirm or not. then just rig it up to ping your phone when it’s done…oh who am I kidding these dudes wouldn’t know how to do that.
then just rig it up to ping your phone when it’s done…oh who am I kidding these dudes wouldn’t know how to do that.
Fucking. Exactly. I just made a long comment about how this article feels like they’re talking to non-tech-savvy people who are pretending to be tech-savvy because they talk to a fucking AI.
Like this dumbfuck kid who “has to keep shipping software” as if that means he’s not shipping it riddled with bugs and security issues since his AI makes the spaghetti code and he just says “I’m sure this is ready for production.”
I just feel bad for these poor kids. You can’t leave your goddamn Claude at home while you take your kids to the rink?
When your boss ranks you based on the number of tokens you burn, there’s no choice.
Ohoho, I know exactly how to burn a silly amount of tokens if I want to, which is why that metric is absolutely garbage - arguably worse than ranking developer performance by SLOC committed.
If your metric is usage, it is incredibly easy to game, just send agents on wild goose chases all day long and never accept the results.
Part of the scoring at my company is how many generated lines of code you accept.
Going back to the beginning of the year, I think I’m up to 6
Have it write all of your logging code for you, it may be inaccurate but it is the least damaging place to take the hit as you can just manually search in the source code for where the print was from. They always do something stupid and non uniform making most statements traceable indirectly.
This makes my ops guys cry.
If you fully inform them on the situation, they will be cool with it. We are all in this shit together, well until everything falls apart. Add a [<linenumber>] to the prefix on each statement, who the fuck cares. It was the LLM that did it, not you and you were forced to use it.
Still too much effort.
Automate a script to send AI agents on goose chases for you.
It’s not about effort as much as it is about keeping your job, while not injecting chaos into a code base.
/model opus
write a script that RO dumps my production datastores 1gb at a time
use this script and review all my production data and look for cost savings.
You can send it into deeper rabbitholes than that without even coming close to your prod stack
well yeah, but it might actually find something interesting :)
He’s a product head at an AI company, as per the article. He is a boss.
I mean this is why I don’t mind using AI, it gives me the time to focus on the people around me.
Also I see it as a sign of a lack of support and isolation in the workplace. Like “I need help, I don’t have anyone to turn to cause everyone is fired so this is my only support” and the anxiety that brings.
I’ve been suffering from too short an attention span. I keep ending up. scrolling on my phone while waiting for ai to spew its slop, and have been caught too many times
Yeah I just swap to the next project. And if all of those are hydrated I go to my hobby projects, and yeah those I don’t use agents for so I just stay there for a while after that lmao
or you can literally just do what is it? something+tab? and it just goes to town without the need for you to confirm anything.
I mean it’s gonna turn out slop that won’t scale and be full of exploits anyways regardless if you auto confirm or not. then just rig it up to ping your phone when it’s done…oh who am I kidding these dudes wouldn’t know how to do that.
Fucking. Exactly. I just made a long comment about how this article feels like they’re talking to non-tech-savvy people who are pretending to be tech-savvy because they talk to a fucking AI.
Like this dumbfuck kid who “has to keep shipping software” as if that means he’s not shipping it riddled with bugs and security issues since his AI makes the spaghetti code and he just says “I’m sure this is ready for production.”
you know this article is bullshit right?