@SenSanders It's not either/or. The best solution for students AND teachers is both.
Teachers can't personalize lessons and pace to individual students. AI can't replace the human element.
Tech for lessons/exercises, with teachers helping guide and motivate, makes everyone better off.
@griswold When we last checked, many did parts well (curriculum design, assessments), but none were set up to be a standalone, student-guided app with dynamic assignments and AI grading.
Much has changed in the last year though, so worth reevaluating. Will post updates!
Writing has to be the most overlooked skill in schools today.
It gets ignored because:
- Itโs harder to grade and test
- Itโs hard to separate from content
But at its core, how you write is how you think. Itโs critical for communication and understanding.
We need to fix this.
Great idea, but gamification for edtech is the easy part. Delivering real learning outcomes is hard.
Duolingo couldn't even get language right and that's its only focus. This would have to do better, for every subject type, on the fly.
Good luck, let's see how this progresses.
I left NotebookLM a few months ago to solve a bigger problem in learning.
Today, as the first step, we are launching @WonderingApp for early access.
It's Duolingo for anything โ turning any topic into a guided path with bite-size visual lessons that can fit into your busy schedule.
But you don't sacrifice depth/effectiveness for convenience:
Total Control: You decide how deep you want to go, how difficult the material should be, and how personalized the experience feels.
Active Learning: We provide the tools you need to practice, test your understanding, and actually apply what youโve learned.
Long-term Mastery: Itโs built to help you truly remember and master any subject, not just skim the surface.
@GergelyOrosz Can't understand why the UI/UX is still so poor and hasn't changed much in these multi billion dollar chat apps.
The closest ChatGPT has to organizing topics is still Projects, which are basically useless. Haven't even bothered to allow threads inside chats...
@redaction Writing is a fun one to try this for.
- Math, you have clear progressions from base skills to compound ones
- Writing, gaps could be due to grammar, vocab, reading comprehension, rhetorical tools, reasoning, etc
Worthwhile endeavor, and we're doing it, but it's tricky.
Odd how hard it is for critics on here to grasp that "not all screen time for kids is equal."
- Watching "educational" Youtube Shorts - bad
- Unstructured iPad time in class on apps - bad
- A structured curriculum, supervised by guides = efficient personalized learning
@KellyClaudeAI@KellyClaudeAI , I saw this video: https://t.co/vdiQrh3OK3 on using DNA results from services like 23andme and cross referencing them with sources to further personalize recommendations (ie. you metabolize _ poorly).
Is this something you think would be valuable to add?
@adamboxer1 Time will tell, but hopefully that's a losing bet.
They're already launching the platform for home schooling using vouchers, which will make it much more accessible than the private schools.
Those of us building apps for them truly believe this is just the start.
The fun part about building apps for students is that your clients are all hackers.
"Okay, made the lesson!
... Now what's every way they're going to try to pass this without actually learning? ๐ค"
AI isnโt wiping out every software job equally.
Edtech, for example.
- Anyone can spin up a โgamified learning app.โ
- Very few can design one that actually drives mastery.
Good outcomes require real instructional expertise and judgment.
That's what we're working on.
@roger_jgs404040 Supposedly anthropic is banning accounts that use their max token to power OpenClaw. The workaround would be getting your AP-key powered "brain" to use Claude Code for tasks.
But agreed, the engine definitely isn't running smoothly.
Finally got OpenClaw running smoothly. Ask it to make a dashboard and it succeeded... in burning $50 worth of tokens.
It's been fun, I'm going back to the Claude Code Max plan. Y'all enjoy refining this one, I'll check up on it again in a month.
@MrZachG Didn't take long to see that:
- AI without learning science != effective outcomes
- Effectively incorporating those is hard, and AI is far from being able to do that on its own
Education isn't solved yet, but we're working on it.