Sabitlenmiş Tweet
[BLOG] The Power BI modeling MCP server in plain English
sqlgene.com/2025/11/19/the…
English
Eugene Meidinger
29.1K posts
Eugene Meidinger
@SQLGene
Data training for busy people. Power BI Consultant. Pluralsight author. He/Him. Mast: @[email protected] Bsky: @sqlgene.com
Pennsylvania, USA Katılım Ocak 2013
627 Takip Edilen3.1K Takipçiler
The problem with being a Boros player is the guilds symbol *really* looks like it could be fascist
English
@Stone_Tao The amount of mocking you can do for external services can get pretty wild.
simonwillison.net/2026/Feb/7/sof…
English
@Stone_Tao I think you'd need pretty aggressive testing and CI/CD
banay.me/dont-waste-you…
English
genuine question. how do you debug code and ensure good quality when coding models spit out 1000s of lines
i still cannot feel comfortable not understanding what every generated line does, reducing the productivity gains coding models should be giving me
Noah@NoahKingJr
Vibe coders debugging an app they built with Claude Code:
English
English
@makeitrhyme9 @WithinRafael They can be removed for violating the NDA, which Microsoft accused that person of doing.
English
Well this is awkward. Recall Microsoft MVP program ejected at least one MVP for helping folks turn on/off these same feature flags. 😂
phantomofearth ☃️@phantomofearth
Microsoft is FINALLY adding a "Feature Flags" page to Windows Insider settings that will let you manually toggle new features on or off without needing to use vivetool or suffering through CFR! (Hidden bits in 26300.8155, not enabled yet)
English
@ciphergoth @EigenGender LLMs have been shockingly good at tip-of-the-tongue tasks in my last experience 😄
English
@SQLGene @EigenGender Claude was able to name it based on your tweet btw!
English
the thing about newcombs problem is that everyone gets hung up on super intelligent predictors but in real life people basically just read each others minds
English
@PawelHuryn How do you know it doesn't change the chain-of-thought tokens and therefore how it thinks?
English
A 16-year-old cut Claude's output tokens by 75%.
The trick: make it talk like a caveman. Less "I'd be happy to help," more "done."
I tested it. Instructions change how Claude talks, not how it thinks.
Example prompt:
"Me talk short. No explain. Tool first. Result first. Me stop. No filler. No polite. Just do."
Om Patel@om_patel5
I taught Claude to talk like a caveman to use 75% less tokens. normal claude: ~180 tokens for a web search task caveman claude: ~45 tokens for the same task "I executed the web search tool" = 8 tokens caveman version: "Tool work" = 2 tokens every single grunt swap saves 6-10 tokens. across a FULL task that's 50-100 tokens saved why does it work? caveman claude doesn't explain itself. it does its task first. gives the result. then stops. no "I'd be happy to help you with that." no "Let me search the web for you" no more unnecessary filler words "result. done. me stop." 50-75% burn reduction with usage limits getting tighter every week this might be the most practical hack out there right now
English
@OneGravitas @EigenGender I mean sure. If you had the time and space I imagine you could properly communicate "An amount of money that would be very annoying to give but you have in the bank".
I feel comfortable assuming that if you are reading this on X, $1k is within an order of magnitude of that.
English
@SQLGene @EigenGender Hypothetical involving money always seem to not really get marginal utility. 1000$ ranges from "more money than they'll see in a year" to "I need to call my personal assistant to handle transactions that small" depending on the person.
English
@ciphergoth @EigenGender THANK YOU! I tried googling based on a vague description and I couldn't find it.
English
A $200 subsidized plan where you can spend up to $2,000-$5,000 of API inference is always going to cause mixed incentives, perverse incentives, and frustration.
English
@RosieCampbell @wfenza @rgblong Elves are terrific. They beget terror.
The thing about words is that meanings can twist just like a snake, and if you want to find snakes look for them behind words that have changed their meaning.
No one ever said elves are nice.
Elves are bad.
-Terry Pratchett
English
@RosieCampbell @wfenza @rgblong Elves are wonderful. They provoke wonder.
Elves are marvellous. They cause marvels.
Elves are fantastic. They create fantasies.
Elves are glamorous. They project glamour.
Elves are enchanting. They weave enchantment. 1/2
English
If someone isn't actually having an acute mental health episode I think we should call it "AI enchantment" rather than "AI psychosis"
(h/t @rgblong)
English
@UsagiKaede1 This came out a few years after cumulative upkeep so I think it was meant for that?
English
This has got to be one of the worst cards ever printed. You get 3 mana but then discard your hand???? What are you going to spend that mana on with no cards???? Unplayable garbage.
English
@jumperz They started integrating Claude a month ago, iirc
pcworld.com/article/308270…
English
interesting how microsoft has copilot and they spent billions embedding it into 365 and now they’re letting claude which is a direct competitor
that tells you microsoft has already decided that Microsoft 365 is their real product and not copilot..
they would rather be the indispensable data layer that every AI reads from than win the model race
think about it this way, they’re trying to be the api layer for the entire AI economy..and that’s smart honestly
Claude@claudeai
Microsoft 365 connectors are now available on every Claude plan. Connect Outlook, OneDrive, and SharePoint to bring your email, docs, and files into the conversation. Get started here: claude.ai/customize/conn…
English
Eugene Meidinger retweetledi
The more I use AI tools, the more I have to admit that I'm not that much more productive... I simply FEEL that much more productive.
In reality, the context switching of kicking several things off wipes out my perceived productivity gains. At least in many/most cases!
English
@Sauers_ Gemini is straight up depressed and self-loathing. Claude is high-fubctionjng anxious
English
GDM should replicate this in Gemini lol
BOOTOSHI 👑@KingBootoshi
you're telling me if we give Claude a hook on every run error that injects the message: "its ok buddy. don't worry about the failure. i think you're doing great" IT WILL PREVENT IT FROM CHEATING? ARE YOU SERIOUS LOL the real agi were the friends we made along the way <3
English
If LLMs are so smart, why do they need all these prompts, harnesses, post-training, scaffolding, etc.?
English
According to steam hardware survey Linux jumped up to 5%+ marketshare.
English