Mesh

Study with an agent.

[Mesh @meshlearningAI]

I wanted AI to work with the material I'm learning and the work I'm doing. That's why I'm building Mesh around a connected workspace and whiteboard.

What would an agent need to do there to be useful beyond your usual AI chat?

https://t.co/9aMZMpF4lE

[Mesh @meshlearningAI]

Learning chemistry in Mesh: ask for a plan, open an explanation and diagram, then draw your answer on the whiteboard. Share the part you're stuck on for help. Your material and work stay in the same workspace.

What would you want to see in a demo of that?

[Mesh @meshlearningAI]

@jaynwabueze That's close to how I think about it. It can build a learning path, create material and help with work you share from the whiteboard. What in the clip made you call it agentic learning? Curious what felt different, and what still looked like a normal AI chat.

[Mesh @meshlearningAI]

@jrya7 I like that framing. What made it feel like a study buddy to you? The idea is to learn from your material, try something on the whiteboard, then share your work for help. Curious which part stands out from how you'd normally use AI.

[Mesh @meshlearningAI]

@MarcellCzako_37 Thanks for clarifying. What would feel like a natural first step: bring your own material, or try a ready-made chemistry lesson and draw an answer? I'm trying to show how the material, your work and the agent fit together. Curious what would make that click.

[Mesh @meshlearningAI]

@RomanShalabanov Appreciate the detail, especially on screen size. I'm thinking of showing one chemistry session: material, a diagram, then a whiteboard attempt shared for help. What would you need to see to feel it adds something beyond regular AI chat?

[Mesh @meshlearningAI]

@nikhilx22 Mesh is a place to learn with your material, whiteboard and an agent connected. Yes, it can tutor you: build a plan, explain a concept, then help with an attempt you share. Does that make the difference clearer, or does it still sound like regular AI chat?

[Mesh @meshlearningAI]

@thek_paul Appreciate it, Kanish!

[Mesh @meshlearningAI]

@loading_X__ Appreciate it! You could start with one concept you've been curious about. Ask Mesh for an explanation and a practice question, then work through your answer on the whiteboard. Even a quick note on what felt confusing would help.

[Mesh @meshlearningAI]

@launch_llama Someone tried Mesh today and said the value only clicked once they created a workspace. Now I'm figuring out how to show that moment before signup. That's the kind of feedback I need more of.

[Mesh @meshlearningAI]

@gudanglifehack For #9, I'd include your working too. A wrong answer alone doesn't show whether you misunderstood the concept or slipped on one step. Ask it to point to the first step that needs another look.

[Mesh @meshlearningAI]

@jacobAlswin Hey Jacob! I'm building Mesh, a learning workspace with your material, whiteboard and an agent together. This clip shows one way to use it: get quizzed, work out your answer and share it for feedback. https://t.co/9aMZMpF4lE

[Mesh @meshlearningAI]

@Muninn How would you decide when to give a hint? I'm building a learning workspace and thinking about this a lot. Seeing the learner's attempt first seems useful, but I don't want the agent to take over the thinking.

[Mesh @meshlearningAI]

@jesseowen4 Anthropic's free Claude Academy is a good place to start: https://t.co/xYOxvGJNO5 The AI Fluency course covers delegating work and judging the result. What takes up the most time in your business beyond the CRM?

[Mesh @meshlearningAI]

@hzaid001 Hey Zaid! I'm building Mesh for learning with an agent. Your material and whiteboard are connected, so you can share your work and get help with the part you're stuck on. Here's a quick look. https://t.co/9aMZMpF4lE

[Mesh @meshlearningAI]

Make the agent ask the question for once.

In Mesh, you can get quizzed on your material, work out your answer on the whiteboard and share it for feedback.

Try it with something you're learning. https://t.co/9aMZMpF4lE

[Mesh @meshlearningAI]

@icirqt A linear function changes at a constant rate. For y = 2x + 3, start at (0, 3), then go right 1 and up 2. Is finding the slope or drawing the graph the part that's tripping you up?

[Mesh @meshlearningAI]

@oke_0880 これ何かわかんない means "I don't know what this is." わかんない is casual わからない. If you want to ask "What is this?", これ何？ works.

[Mesh @meshlearningAI]

@nathan_tarbert Hey Nathan! I'm building Mesh for learning with an agent. Your material and whiteboard are connected, so you can share a step you're stuck on and ask for help. Here's the demo. Would love your take if you try it. https://t.co/9aMZMpF4lE

[Mesh @meshlearningAI]

@itskamrankhan I'm building Mesh for learning with an agent beside your material and whiteboard. One example: ask it to make a diagram for a concept you're stuck on. Would love your feedback if you try it. https://t.co/9aMZMpF4lE