ทวีตที่ปักหมุด
[BLOG] The Power BI modeling MCP server in plain English
sqlgene.com/2025/11/19/the…
English
Eugene Meidinger
28.9K posts
Eugene Meidinger
@SQLGene
Data training for busy people. Power BI Consultant. Pluralsight author. He/Him. Mast: @[email protected] Bsky: @sqlgene.com
Pennsylvania, USA เข้าร่วม Ocak 2013
626 กำลังติดตาม3.1K ผู้ติดตาม
Amusing how much the knowledge economy is being held back right now because we settled on crappy legacy formats like .ppt, .xls, .pdf, .docx and so on
English
@pvergadia JUST DROPPED a month ago 😛.
Also there where situations where it didn't impair learning.
English
JUST DROPPED: Anthropic's research proves AI coding tools are secretly making developers worse.
"AI use impairs conceptual understanding, code reading, and debugging without delivering significant efficiency gains." -- That's the paper's actual conclusion.
17% score drop learning new libraries with AI.
Sub-40% scores when AI wrote everything.
0 measurable speed improvement.
→ Prompting replaces thinking, not just typing
→ Comprehension gaps compound — you ship code you can't debug
→ The productivity illusion hides until something breaks in prod
Here's why this changes everything:
Speed metrics look fine on a dashboard.
Understanding gaps don't show up until a critical failur and when they do the whole team is lost.
Forcing AI adoption for "10x output" is a slow-burning technical debt nobody is measuring.
Full paper: arxiv.org/abs/2601.20245
English
@GaryMarcus Worth noting that this was a 30 minute task and there was a potential model for retaining learning.
English
“AI use impairs conceptual understanding, code reading, and debugging without delivering significant efficiency gains” — and that’s a quote from Anthropic’s own research!
Priyanka Vergadia@pvergadia
JUST DROPPED: Anthropic's research proves AI coding tools are secretly making developers worse. "AI use impairs conceptual understanding, code reading, and debugging without delivering significant efficiency gains." -- That's the paper's actual conclusion. 17% score drop learning new libraries with AI. Sub-40% scores when AI wrote everything. 0 measurable speed improvement. → Prompting replaces thinking, not just typing → Comprehension gaps compound — you ship code you can't debug → The productivity illusion hides until something breaks in prod Here's why this changes everything: Speed metrics look fine on a dashboard. Understanding gaps don't show up until a critical failur and when they do the whole team is lost. Forcing AI adoption for "10x output" is a slow-burning technical debt nobody is measuring. Full paper: arxiv.org/abs/2601.20245
English
AI is still not thinking.
Lossfunk@lossfunk
🚨 Shocking: Frontier LLMs score 85-95% on standard coding benchmarks. We gave them equivalent problems in languages they couldn't have memorized. They collapsed to 0-11%. Presenting EsoLang-Bench. Accepted to the Logical Reasoning and ICBINB workshops at ICLR 2026 🧵
English
declining x axis to trick our mind into seeing “top right good”
Cursor@cursor_ai
Composer 2 is now available in Cursor.
English
@dbreunig The DSpy meetup tonight was fantastic - even though I've never used DSpy, the case studies were full of general insights. Will speakers be sharing their slides?
English
@lossfunk LLMs are *awful* at repeated text. Expecting them to do brainf*ck is silly.
research.trychroma.com/context-rot
English
🚨 Shocking: Frontier LLMs score 85-95% on standard coding benchmarks. We gave them equivalent problems in languages they couldn't have memorized. They collapsed to 0-11%.
Presenting EsoLang-Bench.
Accepted to the Logical Reasoning and ICBINB workshops at ICLR 2026 🧵
English
I am simply not made for git. Switching feature branches, getting the latest changes etc etc...
Most simple way? New folder, start from scratch and dump a local feature branch in it.
Delete old folder with whatever did not want to switch.
English
@fchollet Esolangs is a good test. Brainf*ck is just rude. LLMs are awful ate repeated text.
research.trychroma.com/context-rot
English
This is more evidence that current frontier models remain completely reliant on content-level memorization, as opposed to higher-level generalizable knowledge (such as metalearning knowledge, problem-solving strategies...)
Lossfunk@lossfunk
🚨 Shocking: Frontier LLMs score 85-95% on standard coding benchmarks. We gave them equivalent problems in languages they couldn't have memorized. They collapsed to 0-11%. Presenting EsoLang-Bench. Accepted to the Logical Reasoning and ICBINB workshops at ICLR 2026 🧵
English
@Winterrose You could probably vibe code a browser extension, honestly
English
I would pay $1 per month for life for a browser I could tell my cookie preferences to and never see a f*cking European GDPR popup from ever again.
how did these idiots let this happen
English
@mihaimaruseac True, but I'm not sure how much that distinction changes the negative outcomes or what you have to do to mitigate them.
English
Is AI killing open source?
No. Lazy people that submit slop and don't care about the project, only about their ego, do.
English
The next Slammer word will be a markdown file
Zack Korman@ZackKorman
You can hide these !commands in html comments so people don't see them when reading the skill. The command executes without the AI even knowing about it.
English
SharePoint lists as shortcuts, I’m going to use this, nice feature.
English
Eugene Meidinger รีทวีตแล้ว
At #FabCon and our first-ever #SQLCon in Atlanta with 8,000+ customers, partners & community members.
Enterprises are moving fast to AI-driven, multi-agent systems. The blocker isn’t models. It’s data + context.
New this week in #MicrosoftFabric: Database Hub • OneLake updates • Fabric IQ + integrated planning • Data agents (GA). Get the details: tinyurl.com/5n6k5fhr
English
Should be a $100 deposit if you want to monetize your account
Nikita Bier@nikitabier
The fortress we are building—and the layers of redundancy—to protect the platform against the AI Slopacalypse will seem obvious in a few months. Whether we use every tool in our toolkit is TBD, but it would be negligent to not have them ready.
English