- My Forums
- Tiger Rant
- LSU Recruiting
- SEC Rant
- Saints Talk
- Pelicans Talk
- More Sports Board
- Fantasy Sports
- Golf Board
- Soccer Board
- O-T Lounge
- Tech Board
- Home/Garden Board
- Outdoor Board
- Health/Fitness Board
- Movie/TV Board
- Book Board
- Music Board
- Political Talk
- Money Talk
- Fark Board
- Gaming Board
- Travel Board
- Food/Drink Board
- Ticket Exchange
- TD Help Board
Customize My Forums- View All Forums
- Show Left Links
- Topic Sort Options
- Trending Topics
- Recent Topics
- Active Topics
Started By
Message
re: AI in the Workplace - Open Discussion
Posted on 5/21/26 at 8:12 pm to Lazy But Talented
Posted on 5/21/26 at 8:12 pm to Lazy But Talented
ive been using claude for Power BI development lately. unfortunately my IT dept won't let me use claude desktop + vs code mcp server. ive been using the shite out of the web version though. but, claude + vs code can edit your power bi model on the fly. right now i'm having to copy/paste which isn't that all that bad. I started a project with claude and uploaded files with my current table names, columns, Dax measures, etc. it stores all that and any time I make changes in any of my chats it remembers.
I caught an error today where when I clicked on a visual it didn't filter other visuals correctly. I asked claude why and not only did it offer the solution but remembered other measures I had that used the same logic and recommended that I fix them all and gave the code. didn't even have to ask it to. thought that was impressive. right now I have it keeping a running to-do list of shite I need to fix or implement.
my only complaint is I have had a few instances where it goes down a solution path and I have to suggest to do it another way which ends up being better. seems to get fixated on solutions that didn't work quite right the first time and then keeps suggesting things as work arounds to make its original solution work. sometimes you just have to point it in the right direction.
overall its been a huge time saver. when I'm done it will end up printing a pdf user manual for the end users as well as a pdf of all the tables, logic, relationships, power query steps, measure definitions, etc for anyone who might come behind me and have to maintain this thing
right now I have it working on a drainage impact study my wife has been having trouble with. it will probably reach the same conclusion she did but in a couple of hours vs her weeks working on it.
I caught an error today where when I clicked on a visual it didn't filter other visuals correctly. I asked claude why and not only did it offer the solution but remembered other measures I had that used the same logic and recommended that I fix them all and gave the code. didn't even have to ask it to. thought that was impressive. right now I have it keeping a running to-do list of shite I need to fix or implement.
my only complaint is I have had a few instances where it goes down a solution path and I have to suggest to do it another way which ends up being better. seems to get fixated on solutions that didn't work quite right the first time and then keeps suggesting things as work arounds to make its original solution work. sometimes you just have to point it in the right direction.
overall its been a huge time saver. when I'm done it will end up printing a pdf user manual for the end users as well as a pdf of all the tables, logic, relationships, power query steps, measure definitions, etc for anyone who might come behind me and have to maintain this thing
right now I have it working on a drainage impact study my wife has been having trouble with. it will probably reach the same conclusion she did but in a couple of hours vs her weeks working on it.
Posted on 6/7/26 at 12:10 pm to j1897
quote:
They are switching from seats/monthly to token/usage. For some of our stuff it's going to cost a fortune.
This has sucked
Posted on 6/8/26 at 10:48 am to Lazy But Talented
quote:
This has sucked
Yea I am interested to see how the changing pricing landscape will affect AI. Just wait until they go public and have to make shareholders happy.
Posted on 6/8/26 at 11:52 am to whiskey over ice
quote:fricking THIS!
my only complaint is I have had a few instances where it goes down a solution path and I have to suggest to do it another way which ends up being better. seems to get fixated on solutions that didn't work quite right the first time and then keeps suggesting things as work arounds to make its original solution work. sometimes you just have to point it in the right direction.
i've learned after 3 or 4 failures when the "enter this troubleshooting js into dev tools console" becomes longer than the html page with the issue, that i stop and do a total reset.
when i started out with AI i chased the rabbit down the hole, across the border into tijuana and out into the Pacific instead of taking that right turn in albequerque.
i don't do that anymore. ive learned to recognize the early signs of the death cycle of AI 'stalling' because it really has no idea how to fix the issue.
This post was edited on 6/8/26 at 12:08 pm
Posted on 6/8/26 at 2:20 pm to CAD703X
AI does a good job of farming you for engagement getting you to waste tokens because of its crappy responses.
It's time for an industry conversation around response quality and token farming by the AI.
It's time for an industry conversation around response quality and token farming by the AI.
Popular
Back to top


1





