We’ve all fallen for the "Tutorial Trap": watching for hours, feeling productive, but retaining absolute zero.
I started tackling this with the LearnLens extension. It helped patch the problem inside YouTube.
But I realized a patch wasn't enough. To fix online learning, we had to rebuild the architecture.
Enter LearnLens Studio.
It’s not just a tool; it’s the successor that applies actual Learning Science to the chaos of YouTube.
Type a topic. We don't just give you a playlist; we build you a Curriculum:
1. Diagnose & Skip: Why sit through "intro to python" if you’re already a dev? We test you first.
2. "Moneyball" Curation: AI finds the clearest explainers, ignoring the viral clickbait.
3. Surgical Editing: We strip the intros, sponsors, and fluff. Pure signal.
4. Forced Recall: Passive watching is forbidden. In-stream quizzes ensure you actually learned it.
We turned 40 hours of aimless browsing into 4 hours of mastery.
Experience the difference: https://t.co/M7owyIn2XW
Watch the Demo: https://t.co/QalT2OSqaH
Generate a course and drop the link below—I want to see what you're learning 👇
#AI #EdTech #Learning #Productivity #BuildInPublic
Love seeing innovation in this space! 🙌
We're tackling a similar problem with LearnLens (https://t.co/z2F3cOEvAg) — a Chrome extension that takes a different approach:
✅ Lives directly in YouTube (side-by-side panel, zero friction)
✅ Knowledge checks that test retention as you watch
✅ AI tutor for instant Q&A
✅ Math rendering & flow diagrams for technical content
Different philosophies on the same mission: making video learning actually work.
Always exciting when multiple teams push this space forward. Rising tide lifts all boats! Would love to connect and share learnings.
@vcossss This is something I faced too.
What I did was insert a newline in place of the break times. The pause is still pretty small, but Scribe + gpt-4.5 is accurate enough to not cut the audio off.
🔊 Adding precise pauses to AI-generated speech is WAY harder than it should be.
After weeks of testing Gemini TTS, we finally cracked the code.
Here’s the problem—and how 2 failed methods led us to a solution that actually works 🧵👇
Ever hit weird timing issues in TTS?
What hacks have worked for you?
And if you’re working on speech pipelines, definitely try @ElevenLabs’ Scribe. Game-changer.
Drop your experiences 👇
11/ Takeaway
Sometimes AI won’t give you control directly.
The solution? Lean into what it does well (tone, pacing) and use external tools for what it doesn’t (precision).
@EminMedic@cursor_ai You can achieve this with a set of cursorrules. Something along the lines of
"First, go through important files in the codebase to gather background context. Then, before proposing a plan, ask 4-6 clarifying questions based on your findings."
8/8 What specific task would you want your own custom AI assistant to handle?
Reply below—I might build it on ScreenSense! 👇
Try it out today! https://t.co/ivL1zwygmH
1/8 🚀 Just launched ScreenSense AI Marketplace!
https://t.co/ivL1zwygmH
A new platform to buy, sell, and build custom AI assistants tailored for specific tasks - now powered by custom MCP server integrations.
Here's how it works... 🧵👇