Episode publish date
June 19, 2025 11:01 AM (UTC)
Last edit date
Feb 16, 2026 9:38 PM
Last snip date
February 16, 2026 4:57 PM (UTC)
Last sync date
February 16, 2026 4:58 PM (UTC)
Show
Lenny's Podcast: Product | Career | Growth
Snips
1
Warning
⚠️ Any content within the episode information, snip blocks might be updated or overwritten by Snipd in a future sync. Add your edits or additional notes outside these blocks to keep them safe.
‣
Your snips
‣
[00:00] Prompt Engineering Still Vital
‣
[07:50] Medical Coding Prompt Success
‣
[12:07] Use Few-Shot Prompting
‣
[25:00] Decompose Complex Tasks
‣
[28:42] Use Self-Criticism Technique
‣
[30:10] Provide Additional Context
‣
[01:40] Papers mentioned in the episode
‣
[40:30] Ensemble Multiple Prompts
‣
[46:03] Still Use Chain of Thought Prompts
‣
[52:09] Prompt Injection Persists
‣
[56:32] Agentic AI Security Threat
‣
[01:09:41] Prompts Alone Can't Stop Injection
‣
[01:15:00] Injection Is Unsolvable Problem
‣
[01:20:40] AI Misalignment Risk Grows
‣
[01:24:20] Support Regulation, Not Stoppage
‣