Sebastian Parra retweeted
A toothpaste company has quietly killed the entire market research industry and nobody is talking about it.
Colgate published a paper showing you can predict real purchase intent by simply asking LLMs to roleplay customers.
And this is beyond insane.
If you ask an AI, "Rate this product from 1 to 5," it gives safe, middle-of-the-road garbage.
So researchers invented a method called Semantic Similarity Rating (SSR).
Instead of asking the AI for a number, they asked it to roleplay.
They gave the LLM a demographic profile. They showed it a product concept. And they asked it to write down its raw, unfiltered thoughts.
Then, they used a semantic model to translate those written thoughts into a numerical score.
The results are staggering.
Tested against 57 real personal care product surveys and 9,300 human responses, the synthetic consumers hit 90% of human test-retest reliability.
They perfectly mirrored how different age brackets and income levels react to price changes.
And they provided detailed, qualitative feedback that was deeper and more critical than what actual humans wrote.
This destroys the economics of traditional market research.
You don't need to wait a month to see if a product will sell.
You can simulate 1,000 hyper-targeted customer interviews overnight.
You can A/B test pricing across every demographic instantly.
Sebastian Parra retweeted
This prompt finds things that make you feel kinda gross, like learning you have cockroaches in the walls:
"Clean up stale agent sessions and coding worktrees on my machine. Scan my activity monitor for suspicious processes from apps or packages that I may have installed at one point but no longer use. Audit any skills, MCPs, or agent instructions with exposed credentials or instructions >4months old to review for deletion."
It seems that the name Cursor @cursor_ai is going to be permanently replaced by Grok Build.
Cursor is becoming Grok Build . The internal/nightly builds of the macos desktop version have already been renamed.
20
Sebastian Parra retweeted
I'm opensourcing holospot materials, go crazy, make things holo, free. Link in the thread:
Sebastian Parra retweeted
The older I get, the more I realize that every big win has a small, boring rule underneath it. Mine is four words: “I won’t quit today.” I can quit tomorrow, just not today.
Sebastian Parra retweeted
[vibe-]codevember 2026 - Day 14 - Tension
spite.github.io/codevember-2…
#threejs
Inspired by this post by @kunalabcdani , using Fable 5.2 and Jev by @typesafeai I build my own version of this self serve kiosk app. I made some improvements:
- Pricing for main product and modifications
- Menu
What if we could manipulate UI at the speed of thought?
Built this voice coffee ordering system with @typesafeai Jev. The drink and any updates show up while I’m still mid-sentence, ~200 ms from word to screen.
Feels pretty magical!
3
1
110
There is absolutely nothing wrong with the offer they made to @levelsio . email never says that he recommends it. In fact, he clearly says, “with your own take,” meaning he’s free to express whatever opinion he wants.
Simply saying it’s a repost for an advertising campaign is enough. I don’t know why people are making such a big deal out of this.
1
40
Where are they accepting CV for the CEO position at NVIDIA? I’d like to apply. $NVDA
31
I honestly don't understand people who do nothing but ask for a new model to come out. What more do they need besides Opus 5.5? That model can do everything. Really, what are they missing? It's as if their work and life experience depended on the new model, rather than on what they can actually do with what's already been released.
20
If you're an AI mentor and you recommend that your students use Higgsfield instead of services like Magnific (For begginers) or (Fal. ai or kie. ai for more advanced users), then either you're ignorant and shouldn't be teaching anything, or you're a bastard who recommends the company with the darkest marketing on the planet.
25


















