Ask HN: Are AI "Chatbot Wrappers" ruining EdTech? I'm testing a proactive UX

[Swadesh9422]

Hey everyone,

I’ve been doing customer discovery with CS students learning Data Structures and Algorithms. Right now, every AI tutor in the market is just a reactive chatbox (like ChatGPT next to a code editor).

The problem is, when a student is completely stuck on a logic problem (like Dynamic Programming), they don't even know what to prompt the AI. They just stare at the screen.

I am validating a new UX: A Proactive AI Mentor without a chatbox.

Instead of the user prompting the AI, the AI sits in the background and watches the code editor. It only intervenes via GitHub-style inline comments when a specific event triggers (e.g., they haven't typed in 60 seconds, or they write an O(n^2) loop when it should be O(n)).

Basically, it feels like a Senior Dev looking over your shoulder, rather than a search engine waiting to be asked.

As developers and founders, do you think this "event-driven/proactive" UX is the future for highly technical learning, or am I overcomplicating it? Would love to hear your thoughts.

[a_void_sky]

I believe it is the way it is because it would cost too much otherwise. You can use the live API for continuously monitoring what the user is doing and if they are stuck but that would be expensive.

[Swadesh9422]

You are right. Streaming every keystroke via a Live API would bankrupt me on day one. My current thought is to fake the 'live' feel using client-side heuristics. I'd only ping the LLM on specific events: a 60-second typing idle timeout, or intercepting the payload when they hit 'Run' and fail a test case. Do you think relying on frontend triggers like that is enough to make the UX feel proactive?

[shubhamintech]

while i was learning, i used to experiment by first solving the problem in terms of input output and then optimise it one by one. and with the technique/triggers you mentioned, i'm not sure if it'd cater to somebody like me