Becoming Dauntless
When I start research for a new tech product, whether that’s an AR/VR app for the Dauntless portfolio or a mobile app for creator co-founder clients, I usually start with the same prompt.
It’s easily the highest-value AI input I use, and it has saved me countless hours building minimum viable products (MVPs) no one wants. It works for tech products, physical products, digital products, consulting, and service offerings. I use Perplexity to run this because it is the best at research and at scraping Reddit forums and social media. You can use those features on the free tier, but you can only run this level of prompt twice in a day before you run out of free credits. Here’s the template, fill in the blanks based on what you want to build.
I'm creating a [app/web app/course/PDF guide/eBook/done-for-you service, etc], and want to base it directly on real frustrations and pain points. Analyze Reddit threads, online discussions, comments sections, new articles, blogs, and social media posts related to [specific customer] who wants to [describe dream outcome]. Look at frustrations, struggles, and frequently asked questions people post. present the response in a table with three columns:
Pain Point / Questions from online forums
Insight / Why it matters
Proposed solution (offering/feature + short description)
Here’s an example prompt and prompt output:
I'm creating a course or digital download, such as a PDF guide or workbook, and want to base it directly on real frustrations and pain points. Analyze Reddit threads and online discussions related to people who want to leave their corporate jobs to work for themselves or start their own company. Look at frustrations, struggles, and frequently asked questions people post. Use these insights to draft a five-module course/book that directly addresses these frustrations. present the response in a table with three columns:
Pain Point / Questions from online forums, Insight / Why it matters Proposed course module/workbook chapter (lesson title + short description)
The output is a meta-analysis of your market. You can take the output and give it to Perplexity Computer or Claude so it can build a product for you, but I’ve found those outputs bland and AI sloppy. For better results: prompt your brain, ask more questions, talk to some humans, and then prompt piece by piece if you're vibe coding.
From here, I can decide what makes the cut for MVP based on what is most likely to get people to download/buy/use the product (aka validate product-market fit). Remember: it’s a prompt, not a magic spell to bypass creativity and product development.
I’ve yet to meet someone in real life who has prompted their way to millions, but iterations of this prompt have at least saved me hundreds of hours, and given us usable data to inform better innovation and creativity. Let me know what you use it to research, and if a workshop or live event to run these in real time would be something you'd sign up for.
Happy Prompting.
xoxo
Lori-Lee
Articles Our Team Slacked each Other This Week
We are living in a simulation, and it has a soundtrack now. Georgia Tech researchers are arranging music to help you see what’s behind you. The same way horror movie music tells you the murderer is sneaking up behind a victim, a two-note Jaws-adjacent motif can alert you that a delivery robot is around the corner and in which direction.
AWE happened last week, here's a recap. (Forbes). We are seeing more consumer headsets at enterprise prices. Which we know is not a recipe for success... But that aside, Forbes called out some cool stuff like a metaverse equivalent of HTTP, so you can browse spatial computing (much needed but not historically profitable), and eye tracking tech to give insights on things like physiological state that developers can glean from various headsets’ biometric APIs.
NVIDIA launched XR AI at AWE 2026. We've had this on our radar for a while. XR AI is a toolkit that developers can use to build AI agents for smart glasses.
Ford Re-Hired "Grey-Beard" Engineers that AI Couldn't Replace. But not to go back to their old jobs, and instead to train younger staff and reprogram the AI tools that missed major quality issues. Turns out you need to train the 30 years of experience into the AI before you can see scalable gains. Shocking.
What Happened At Dauntless HQ This Week
We wrote a proposal for a reverse engineering in a rapid manufacturing supply chain (using our spatial scanning and automated 3D model clean-up, of course). The sci-fi movie inspo? Paycheck from 2003 starring Ben Affleck and Uma Thurman.
The posts, memes, and videos that caught my attention this week:
Opportunities Founders & Entrepreneurs Should Know About
The Small Business Expo is looking for speakers. They have expos across the US and need speakers across business functions.
The Android XR Catalyst accelerator is accepting applications until June 30th. The accelerator provides funding for people building XREAL's Project Aura.
Overheard on Slack
Enter the chat. This week’s best one-liners. No context.
“It’s not pick your player, it’s pick your timeline”
Enjoyed this newsletter? Make sure you are subscribed and forward it to your smartest friend.
