28 Comments
User's avatar
Sharyph's avatar

This sounds like an absolute advantage for leaders.

Moving from "gather/process/analyze" to simply "prompt/review/apply" is a massive shift in how strategic work gets done. It moves you out of the weeds and back into leading.

Expand full comment
Joel Salinas's avatar

100%! It’s the kind of advantage that was only available for those who could hire multiple assistants

Expand full comment
Naina Chaturvedi's avatar

++ Good Post, Also, start here 100+ Most Asked ML System Design Case Studies and LLM System Design

https://open.substack.com/pub/naina0405/p/bookmark-most-asked-ml-system-design?r=14q3sp&utm_campaign=post&utm_medium=web&showWelcomeOnShare=false

Expand full comment
Nitin Sharma's avatar

Love the ‘research collapse’ framework.

This is exactly how I use ChatGPT Agent too, not to do the thinking for me, but to remove the 80% of work that prevents me from thinking clearly in the first place.

Thanks for writing such a valuable post.

Expand full comment
Joel Salinas's avatar

Thanks, @Nitin! I imagined you were doing something similar!

Expand full comment
Hodman Murad's avatar

It's like having a junior analyst who does all the grunt work for me. 🥰 Suddenly, I'm free to actually think about what it all means instead of just drowning in tabs.

Expand full comment
Joel Salinas's avatar

Yes!!! It really is mind blowing how much time we can get back

Expand full comment
Doland White's avatar

Great insights and tips here 🙏

Expand full comment
Joel Salinas's avatar

Thanks, Doland!

Expand full comment
Colette Molteni's avatar

Saving time on busy work frees up time to engage with your team and to spend heads-down time being strategic. Great tips here for Agent prompts and real time savers. Thank you, Joel!

Expand full comment
Joel Salinas's avatar

I have so many automated GPT agent prompts running weekly, haha, but for exactly that, gives me more time for the uniquely human pieces

Expand full comment
Aaqiel Pillay's avatar

Great read Joel! Thanks…but I’d like to say…

Leaders gain more clarity by letting AI do the work, not just advise on it.

Expand full comment
Joel Salinas's avatar

That’s it!

Expand full comment
Epsilon Protocol's avatar

Brilliant breakdown! The shift from gathering data to strategic prompting is exactly what leaders need. Your practical examples make Agent feel accessible, not overwhelming. Thanks for the roadmap!

Expand full comment
Joel Salinas's avatar

Really glad to hear that :) thank you

Expand full comment
Chintan Zalani's avatar

This is really cool. For YouTube, I think I understand that probably browsing manually will uncover more videos. But for a website, I am trying to understand if the deep research agent they have would probably do a similar job to what you describe. Did you find some limitations when you try to use that?

Expand full comment
Joel Salinas's avatar

Some websites like Levi jeans stop agents from accessing, that’s the main one. Deep research works for me when I don’t have set sources, but when I want to follow a certain number of channels in YouTube or Reddit, and want to control those sources, agent works best

Expand full comment
Chintan Zalani's avatar

That makes sense. Thanks for sharing. Yeah there’s definitely a restriction some websites place. Makes sense on controlling sources too!

Expand full comment
Joel Salinas's avatar

Especially since most of ChatGPT deep research is Reddit haha

Expand full comment
Mia Kiraki 🎭's avatar

This is great and timely because I'm trying to "leverage" (hate this word but it's morning and nothing else comes to mind) YT more. Looks like it'd be super worth it for my content strategy. So thank you!!! ❤️

Expand full comment
Joel Salinas's avatar

For sure!

Expand full comment
John Brewton's avatar

Love how this reframes AI from a novelty into a genuine operating advantage.

Expand full comment
Dennis Berry's avatar

This is such a game changer. Thanks for this awesome resource.

Expand full comment
Data Frank's avatar

This is brilliant.

I’ve noticed that most people lose clarity not from lack of tools, but from scattered systems. The way you use Agent feels like structure meeting creativity in real time.

It makes me curious — how do you decide what research deserves automation versus reflection?

Expand full comment
Joel Salinas's avatar

Great question! To me, automation only comes in once there has already been reflection on my part and I am sure of how to articulate the automation needs clearly.

Expand full comment
Suhrab Khan's avatar

This is a masterclass in turning AI into strategic leverage. Leaders who implement these Agent-driven research prompts don’t just save hours, they gain foresight and clarity that manual methods can’t match. The 12-prompt library is the blueprint for anyone serious about leading with intelligence rather than reacting to noise.

Expand full comment
Joel Salinas's avatar

Suhrab, I appreciate that!

Expand full comment
User's avatar
Comment deleted
6d
Comment deleted
Expand full comment
Joel Salinas's avatar

Thanks!

Expand full comment