UJ retweetledi
UJ
546 posts

UJ
@uj_alpha06
I think I know some shit about Thumbnails, Video-Editing, Podcasting & Storytelling
Katılım Ağustos 2016
306 Takip Edilen115 Takipçiler

You spent more time choosing your lunch today than researching your next podcast guest.
Then you wonder why your interview sounds like every other interview
Here's the truth asking the right questions and proper podcast packaging gets you views. Winging it keeps you stuck at the 500-view mark.
But that level of preparation takes 7+ hours per episode. Who has that time?
We're building an app that gives you 7-hour quality in 5 minutes.
Join the waitlist → Like and retweet 👇
English

Stop lying to yourself.
Your podcast isn't failing because of budget
It's failing because you won't do the research.
While you use ChatGPT for 10 minutes shows like @LewisHowes @ChrisWillx spend hours researching their guests.
Real pre-production means:
→ Watch previous interviews of their guests
→ Find unique angles in their expertise
→ Create questions that draw out raw, authentic responses
This process takes 5-6 hours per episode.
My co-founder (@jayyygupta ) and I are building an app that automates this research in 5 minutes.
Like and retweet to join our waitlist
English

I see podcasters getting 847 views and wondering why, so let me explain what’s really happening here.
The truth is you’re asking surface-level questions and your titles aren’t grabbing anyone. I know that stings but stay with me.
Here’s what top podcasters’ teams do that you don’t see. They spend over 7+ researching each guest before the host even knows who’s coming, compiling deep-dive briefs, crafting 10+ title options, and reverse-engineering every question to spark a viral moment.
Obviously you don’t have their setup.
But what if you could compress that entire 7+ research process into just 5 mins?
Me and My co-founder (@jayyygupta ) are building an AI app that does exactly this.
English

Imagine if @MrBeast started a podcast in 2025.
His obsession: 100 million views on episode one.
He'd spend AT LEAST 7+ hours in pre-production PER GUEST.
Here's exactly what his pre-production would look like (and how we're trying to automate it) 🧵
1/ The Digital Archaeology Phase 🔍
MrBeast would become a forensic investigator of his guest.
1. Deep-dive their Twitter, LinkedIn, every digital footprint
2. Extract transcripts from their 10 most viral appearances
3. Map what stories exploded vs died
This alone would take atleast 3+ hours.
2/ The Virality Venn Diagram ⚡
MrBeast's secret sauce would be finding the golden overlap:
1. What makes HIS audience obsess
2. Guest's stories they've never told
3. Patterns from viral podcast moments
Example: If interviewing a founder, he'd find the ONE failure story that matches his audience's biggest fear.
3/ Emotional Architecture, Not Questions 🎯
MrBeast wouldn't write generic questions.
MrBeast would engineer emotional journeys:
1. Study Top podcaster's viral patterns
2. Analyze Top 10 viral moments in guest's niche
3. Build questions that guarantee goosebumps
Not "tell me about your success?" But "Walk me through the day you almost quit?"
4/ The Thumbnail Moment Hunt 📸
Before hitting record, MrBeast would know:
1. The exact story that becomes the thumbnail
2. Which question triggers the breakthrough
Title that creates instant FOMO: Words that tap into primal desires, fears, or the terror of missing life-changing information.
This is how you have a chance to hit 100M views.
5/ Here's the plot twist:
Me and my cofounder (@jayyygupta ) built software that does EXACTLY this!
1. Excavates every guest appearance
2. Maps viral patterns automatically
3. Finds story intersections
Generates 14 emotion-triggering questions
What took us 7+ hours now takes 5 minutes.
We're opening 50 beta spots next week.
Like ♥️ + RT this thread to enter 🔄
The future of podcasting starts with pre-production!

English

Most podcast apps help you edit or transcribe.
Nobody's solving the REAL time sink: pre-production.
Me and my co-founder (@jayyygupta) built an AI in 48 hours that generates your entire podcast blueprint: guest research, questions, YouTube strategy in one click.
Here's exactly how it works 🧵
We've been the research guys and strategists for podcasters for 3 years. We ARE the team they hire.
Every podcast requires:
→ 3-4 hours of guest research (watch their podcasts on different channels to not ask the same questions, search their socials, read their blogs/posts for insights)
→ 2-3 team members
→ Jumping between ChatGPT, Claude, Google
→ Still missing key insights
This is insane for something we do BEFORE even recording.
So we built something different.
THE PODCAST FLOW ENGINE 🚀
It analyzes your guest and generates:
→ Viewer avatar (who's actually watching this)
→ The "big promise" (why they should care)
→ 14-15 questions mapped to the guest's expertise
→ Conversation structure that creates truly authentic and real moments
But here's the game changer:
THE PACKAGING AI 📦
It scrapes YouTube to find what's working RIGHT NOW.
Then generates:
→ 5 title options based on viral patterns
→ Thumbnail concepts that match top performers
→ All specific to YOUR guest + topic
No more guessing what converts.
How it actually works:
→ Enter guest name
→ Add their Twitter/LinkedIn/Instagram
→Add insights about the guest or angles/topics you'd like to explore
→ Hit generate!
In 90-120 seconds you get a complete blueprint that would take a research team half a day.
The tech stack for nerds:
Instead of making you prompt ChatGPT 47 times, we chained multiple AI models:
→ Guest analysis (their work, expertise, controversies)
→ Audience mapping (who cares about this topic)
→ Question generation (mapped to emotional peaks)
→ YouTube optimization (title/thumbnail patterns)
One click runs all of it.
We're not trying to replace human creativity.
We're replacing the repetitive 9-5 research grind that keeps us from the REAL work growing podcasts, building relationships, creating magic.
Why waste talent on tasks AI can handle when we could be focusing on strategy and growth?
This lets us be producers, not just researchers.
Looking for 50 beta testers who:
→ Ship podcasts regularly (not "planning to start")
→ Interview guests (not solo shows)
→ Want to 10x their output without burning out
Comment 🎙️ + your show name
First 50 get free access + direct input on features
This isn't funded. No VCs. No exit strategy.
Just two guys who've been in the podcasting trenches, leveraging AI to build something that actually works.
(Though we are applying to YC Spring 2025 👀)
If you're tired of choosing between great research or having a life, let's fix this.
Who's in?
P.S. Huge shoutout to @mattyp @gregisenberg for the inspiration to build in public, and @Replit for making it possible to ship this in 48 hours, literally built the entire thing on Replit 🔥

English
UJ retweetledi
UJ retweetledi

@cutthepod @calum_johnson9 @SahilBloom I like this concept! I found the Graphics Metric particularly interesting and creative.
It’d be great to see a Storytelling/Value per second Metric too, sharing your thoughts on the podcast’s value and the most engaging topics or stories.
English

🧑🍳 Pod Roast: @calum_johnson9, The Calum Johnson Show w/ guest: @SahilBloom
This is a banger podcast, take notes on how to unf^*k your own pod (and life).
Audio Quality: 1.8/2
• Music is a touch heavy on volume, nothing alarming. Feels fitting to Calum's friendliness.
• Mics are incredible.
Video Quality: 1.9/2
• Love the framing to allow room for graphics on screen uninterrupted. Made possible w/ either 4K or foresight, either way A+.
Editing: 1.6/2
• Yes, some pods have 5x more complex edits. That does not mean they are better. This is a simple approach, done well.
• I would argue there is room for more 'brand/personality'. The film transitions are fine, but feel too stock for such a high quality pod.
Graphics: 1.6/2
• I feel the captions are a touch off-brand. I'd like to borrow from the look of the end-graphic and fat yellow squares on thumbnails to emphasize text vs. this neon approach. But, again they are consistent and intentional therefore, I'm a fan.
Flow: 2/2
Overall Social-Bite:
This is such an effective template to copy for any video podcast out there. You don't have to bring in a metric-ton of effects and b-roll. You can stay simple, deliver a great interview, and have a banger podcast.

English
UJ retweetledi

I will be giving away 100,000 dollars to somebody from my email list on Christmas day.
Rules.
1) You must be signed up to my email list to win, I will randomly select an email address and email the winner on Xmas day.
Sign up for free on cobratate.com/newsletter.
2) You must retweet this tweet. If you are selected, and you didnt retweet, I will choose someone else.
Top G.
The G stands for Generous.
GOOD LUCK!
Andrew Tate@Cobratate
20,000 retweets and I’ll do this again for Christmas.
English

@khemaridh @calum_johnson9 Thanks alot man I appreciate it ❤️ Would love to work with you for your youtube channel
English

Check out this Podcast Intro I recently edited for @calum_johnson9 🚀🚀
DM me if you want to work together!
♥️Likes & Retweets♻️Are Always Appreciated🙌
English


🎬 - A recent Short I edited for @calum_johnson9 & @APompliano Podcast!
DM me if you want to work together!
♥️Likes & Retweets ♻️Are Always Appreciated🙌
English

🎬 - Another Short form video I edited!
DM me if you want to work together!
♥️Likes & Retweets♻️Are Always Appreciated 🙌
#videoediting
English














