Evidence receipt / preference
Published · transcript-backedLenny Rachitsky: preference
19 Jun 2025 Lenny's Podcast AI prompt engineering in 2025: What works and what doesn’t | Sander Schulhoff (Learn Prompting, HackAPrompt)
“Talk about maybe, as we go through them, which one this is most helpful for because it's not just like, "Oh, cool, I'm just going to get a better answer from ChatGPT.”
Source trail
Everything needed to verify it.
- Speaker
- Lenny Rachitsky
- Attribution
- Verified speaker
- Claim type
- preference
- Recorded
- 19 Jun 2025
- Publisher
- Lenny's Podcast
Transcript context
…Exactly. ... achieve the result that they're achieving? Or in Bolt and Lovable. You have a prompt that you give say, Bolt, Lovable, Replit, v0, and then it's using its own very nuanced long, I imagine, prompt that delivers the results. And so I think that's a really important point as we talk through these techniques. Talk about maybe, as we go through them, which one this is most helpful for because it's not just like, "Oh, cool, I'm just going to get a better answer from ChatGPT. " There's a lot more value to be found here. Yeah, absolutely, and most of the research is on those, I guess, now you've coined it as product-focused prompt engineering.…
Stored transcript either side of the excerpt. The highlighted words are the published quote; the surrounding text is unedited source, never generated.