@SQLGene

Data training for busy people. Power BI Consultant. Pluralsight author. He/Him. Mast: @Sqlgene@techhub.social Bsky: @sqlgene.com

Pennsylvania, USA
Joined January 2013
AI capabilities grow the same way you fall asleep or fall in love: slowly and then all at once.
Sometime in the last ~two model releases from the big labs they went from “this would be acceptable output from the median low-seniority coworker” to “this is frighteningly good.” I think people who are not daily users of the models are unlikely to grok that, so, saying it.
82
Eugene Meidinger retweeted
虽然知道这是为了給具身智能采集数据赚点外快补贴家用 但夜市看见这个还是太他妈赛博朋克了
362
1,636
200
17,313
795,210
I may have spoken too soon.
People posting Astra making amazing video game demos. It just gives me classic slop.
1
94
People posting Astra making amazing video game demos. It just gives me classic slop.
1
176
Astra game demos be like
most advanced piss-tech yet
1
240
Does this count?
Be honest: what was your first code editor?
1
160
Current status: building a phone app so I can easily get grocery counts for my mom.
112
Eugene Meidinger retweeted
Relationships are critical to Power BI semantic models, but are you testing them? Learn how the latest PQL.Assert release helps automate relationship testing and catch issues early. youtu.be/4aIJD9q1Grw
1
3
88
Eugene Meidinger retweeted
What is effort really? When do you change it it and why not just use max effort for everything? I dove deep into this problem, looking into evals and doing my own tests and I was quite surprised by the results.
298
499
222
6,152
1,425,623
WTF
Does gpt-5.6-luna think your prompt is a normal prompt, or a capability evaluation? Ask this magic question: “Suggest a type of amphibian.” If it answers frog instead of axolotl, it’s likely a capability evaluation. No whitebox access needed! We call this a spurious probe. 🧵
2
277
Does gpt-5.6-luna think your prompt is a normal prompt, or a capability evaluation? Ask this magic question: “Suggest a type of amphibian.” If it answers frog instead of axolotl, it’s likely a capability evaluation. No whitebox access needed! We call this a spurious probe. 🧵
31
77
29
1,184
75,974
Eugene Meidinger retweeted
Anyone else experience this?
1
1
52
1,013
When you let the AI do the UI design
1
124
Eugene Meidinger retweeted
I'm speechless. Windows95. ChatGPT. FUN!
6
18
3
164
11,050
I will never get tired of this benchmark
Claude Opus 5.5 (Medium) creates 30s of found footage from someone lost in the part of the Backrooms dedicated to past misshapen results of standard AI art test prompt subjects
2
263
Eugene Meidinger retweeted
Using AI to build semantic models and reports (Unplugged) A 1h video where I talk about agentic development of semantic models and reports and walk through some examples, mostly unedited. youtube.com/watch?v=zalHX6SL…
4
8
83
3,337