AI prompt engineering in 2025: What works and what doesn’t | Sander Schulhoff (Learn Prompting, HackAPrompt)

In this episode of Lenny's Podcast, host Lenny interviews Sander Schulhoff, an expert in prompt engineering and AI red teaming. They discuss various prompt engineering techniques, including few-shot prompting, decomposition, self-criticism, and providing additional information. Schulhoff also debunks common misconceptions, such as the effectiveness of role prompting and the use of threats or rewards. The conversation shifts to AI red teaming, focusing on prompt injection techniques and the challenges of securing AI agents against malicious prompts. Schulhoff emphasizes that prompt injection is an ongoing issue, not a solvable problem, and highlights the importance of AI research labs developing robust defenses.