The sudden rise of “vibe coding”—non-programmers using savvy prompting to tease code out of LLMs—has sparked a revolution far beyond computer screens. Now, just about any activity can be tackled simply by harnessing the power of vibes.
“Vibe flying” was the first trend to take off (and, no surprise, first to crash). Then there was the rise of “vibe welding,” quickly followed by an uptick in “vibe firefighting.” Meanwhile, the number of new “vibe surgeons” and “vibe dentists” has been strangely equal to the number of new “real lawyers.” |
|
|
Feel free to run a vibe check (and a check of today’s date) before believing any of that—but the truth is that LLM-assisted coding is here to stay. And while AI alone won’t turn newbies into hackers—knowing just a few powerful concepts can give your vibes a serious supercharge.
In just a few minutes a day—about the time it takes to write a LLM prompt—you can learn things like how to harness the power of functions. Next thing you know, you’ll be building modular, scalable programs. How’s that for good vibes?
|
|
|
🗣 ️Automatones. Prompt these text-to-speech bots to talk however you want—from cowboy, to DJ, to cowboy DJ. The voices are endless. |
🦅 Wing stops. Is watching the birds at your feeder not cutting it? Track the real-time flights of millions with this avian traffic report. |
🟥 “I’ll be block.” One day, Skynet-level AGI will beat humans at this grid puzzle. Until then, enjoy being better than the bots at something. |
⏰ 59 BCE (Before Clint Eastwood). Put the “no” in Anno Domini by measuring dates from the moment of various milestones. |
|
|
Supercharge your vibe coding skills |
Obscure languages and syntax errors are so over. Learn how to harness the power of AI to write code—and have it actually work. |
|
|
Just in time for April Fools’ Day, the social media site Scrollr has added an AI fact-checking feature. Initially, 2/3 of posts are tagged “Real” and 1/3 “Prank” by human reviewers. The AI fact-checker works like so: 3/4 of the time it flips a post’s tag from “Real” to “Prank” or vice versa, and 1/4 of the time it decides no action is needed, which locks the post’s tag forever. If the tag gets flipped, the AI fact-checker does another review.
What is the probability a post will end up tagged “Real”?
Bonus: Scrollr is considering capping the number of revisions a post’s tag can receive. What should Scrollr set as the cap in order to get the number of “Real” and “Prank” posts as close as possible?
|
We’ll randomly choose one correct respondent to each puzzle for a shout-out in next month’s email. |
|
|
Congrats to last month’s winner, Charlie Wang, who solved both puzzles and got a tight approximation for the bonus puzzle.
The key to figuring out which word had the advantage was thinking about how their shared subword (“TALE”) allowed STALE to steal victories out from under TALES. For more details, see this month’s solution, which includes Charlie’s slick argument.
Check out the full solution here. |
|
|
|
Ready to unlock the full Brilliant experience? Go Premium
|
|
|
|
Ready to unlock the full Brilliant experience? Go Premium
|
|
|
|
|
“The fool doth think he is wise, but the wise man knows himself to be a fool.” |
|
|
Download the Brilliant app
|
|
|
|
2261 Market St Suite 4281, San Francisco CA 94114
Unsubscribe from receiving Brilliant emails like this.
|
|
|
|
|
|