Full Post w/ Prompt Pack: https://natesnewsletter.substack.com/p/ai-agents-better-communicator?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true __________________________________ What's really happening with prompt engineering now that AI agents are 100x more powerful than they were six months ago? The common story is that prompt engineering is dead and you can just ask AI for what you want — but the reality is more complicated. In this video, I share the inside scoop on the shift from prompting to questioning AI agents: • Why prompt engineering is now table stakes, not a skill • How to treat frontier LLMs like a senior partner • What the AI Question Method looks like in practice • Where most knowledge workers are still stuck in 2025 habits The risk is that anyone running heavy knowledge work with weak questions gets shallow output from powerful agents, but the operators who learn to ask sharp, layered questions unlock real leverage. Chapters: 00:00 Prompt engineering is now table stakes 01:24 Why agents are 100x more powerful in 2026 02:48 Introducing the AI Question Method 04:05 AI as senior partner, not junior teammate 06:20 Why most people are still prompting like it's 2025 08:10 Defining agents vs. agentic pipelines 10:05 Principle 1: The flashlight intent 12:30 Conveying perspective and edges in your questions 14:45 Principle 2: Asking what good looks like 16:20 The PRFAQ example with Prime Video 19:10 Principle 3: Wrestling with data and opinions 21:30 The MRR product-led growth example 23:45 Why memory and quick-start guides matter Subscribe for daily AI strategy and news. For deeper playbooks and analysis: https://natesnewsletter.substack.com/ Listen to this video as a podcast. - Spotify: https://open.spotify.com/show/0gkFdjd1wptEKJKLu9LbZ4 - Apple Podcasts: https://podcasts.apple.com/us/podcast/ai-news-strategy-daily-with-nate-b-jones/id1877109372
ADVERTISEMENT
It's all context engineering now
i knew this from DAY ONE. Its all about FRAMEWORK, baby!
From my experience, this semi-open questioning technique was great since thinking models came, but yes, now it does miracles with 5.5 and 4.7
Another metaphor might be useful; I've never hd a good manager. BUT: I spent a whole career working with people (either colleagues, clients, or my doctors) who knew way more than I did, about many things. The "question prompting" you describe is exactly the discussions I have with such folk: "My target is this problem A, I think the outer boundaries of the scope are these B to C, I've come to you because I believe you know a lot about D, first, what do you think of this framing? We can take it from there.". Yes, it's how you channel a senior mentor, but also how you collaborate with a very knowledgable peer.
This is honestly one of the most important videos for people! My background in psychology and therapist training helps so much with prompting and getting the AI to actually research alongside me and creating innovative thesis’s. I’m literally sitting in a restaurant one phone this video and the other I’m having GTP5.5 on my macmini at home create a massive research essay for an idea that popped into my head haha. Taste and curiosity are the real power skills in this AI age.
Full Post w/ Prompt Pack: /p/ai-agents-better-communicator?r=1z4sm5&utm_campaign=post&utm_medium=web&showWelcomeOnShare=true
This video is long overdue. Thanks Nate. Oh and BTW the flashlight metaphor…it’s focused beam vs splash.
I’m a 49ers fan but love the old school Seahawks jerseys.
With Opus 4.7, I've begun having 'conversations' with it as if it were just another person. I almost (but not quite :P) forget that I'm conversing with AI.
This is gold. I have always felt the language model’s biggest limitation is my ability to ask the right questions. Wispr Flow has been critical for me to engage with Claude this way.
I'm doing almost as much work on my 5.5+Hermes setup on the $20 plan as I'm doing on 4.7 max 200. And starting to offload tasks to local models under qwen 3.6
Thinking about AI as managing an employee who is smarter than you are rather than an executive assistant who handles your calendar is tight. As a manager in a real company you need to have different styles for those two scenarios. That's the difference between AI six months ago and now.
I agree with you, Nate that we need to ask better questions but how do we ensure the agents have the context and knowledge to answer us legitimately? By asking about the Mona Lisa later in life are we sure the agent knows about the early life cause it will lie to us no problem and will call it hallucination not a lie. How do we solve for that?
I think I just realized that if YouTube is your source of information and guidance, you're already building behind the curve.
Dude loves endless talks. Looks like a clown.
Gold i have just had an argument about this with our ai project leader. This is gold Nate.
Thanks for sharing that. It’s been a while that I treat my agent as Sr Developer companion, asking honest question and for push backs. I try to have very intentional conversation, so the agent keep the right focus. So far I’m happy with the results
That’s exactly what I found works best today 4:39
Go Seahawks! This guy looks just like my neighbor and I'm up in the Seattle area... 🤔🤔🤔
Great insight. I’ve been talking to AI like I do with humans since starting my project. They’re built with mood detection, like talking with colleague’s respect AI enough to paint the picture you want them to see.