What's the benefit of this over, for example, MagicSchool which my school is continually pushing on us and is free?
This next bit is going to sound a bit harsh or confrontational, but trust I'm coming at it with a good heart of hope. I'm a bit on edge about pushing AI for grading. Teachers are... not technically literate folk. They won't understand the nuance of AI and using it to grade student work seems like heartache and disaster in the making. Are you doing anything to somehow subvert the fundamental nature of LLMs to make it suitable for grading?
Not OP, but I work on AI in higher ed at a major university.
I get the concerns about AI grading. The solution isn't to have AI grade entire assignments at once. Instead, break down the assessment into smaller, discrete tasks and develop a grading rubric around those. The idea is to limit how the AI can respond - usually to simple binary choices like completed/not completed, true/false, etc. (Also, the models have been RLHF’d to generally put a positive spin on things, so if anything they’re likely to be overly generous in assessment.)
From there, provide the AI with the answer key, student response, rubric, and any other necessary context then use the Structured Outputs API to force consistent responses for each discrete task. I've had the most success using boolean values or simple enums (like "Correct", "Partially Correct", "Incorrect"). You can include a field for reasoning, then chain AI calls to get a second assessment as verification.
I really hate to rain on people's parade or discourage them, but honestly - this is not only incredibly low-effort AI slop but dangerous low-effort AI slop.
It appears you did little or possibly no research into existing tools, talked to no actual teachers about their needs, or even put any thought into it. An AI slop generater itself generated by AI, with no reference to market needs ... and you expect people to use it to educate children?
Oh that happens all the time. Right up front, I would not have gotten through my first year teaching if I had to write all my own lessons. My whole first semester I just copied lock step what the other subject teacher was doing. If I didn't have her, I would not have survived.
To be clear, though, that doesn't change by adding AI. If I only had AI, I would also have drowned and failed miserably. And that's despite knowing my content forward, backwards, and sideways.
The few times I've tried generating lessons from AI, they've been boring and uninspired. With lessons, it works best in reverse, taking a human-generated lesson and stuffing it into a easier-to-share and easier-to-tweak lesson plan format.
Yet another "AI sign up with Google web site with its pricing page" that will fuck up young people. I’m fed up with this. Flagged. Go away. Sorry for the rant.
> product creation grind
You’re the new SEO spammer that we had ten years ago except it’s AI this time.
I’ve been on a product creation grind, pushing out something new almost every week, but this time I’ve got something that’s been in the works for a bit longer.
I wanted to get your thoughts on my latest launch and hear any recommendations you might have. Not trying to market here (I’m guessing there aren’t too many teachers around), just looking for advice and feedback on my product.
As usual, I’m starting with a Product Hunt launch to test the waters. It’s going “alright” so far I’m currently sitting at #5 today, but I’m hoping to climb just a little higher.
I'm not a teacher, but has worked as a tutor for several years. Interesting thing you have there.
Just a few thoughts on that:
Classroom assignments are the worst nightmare! If you could solve this, it would be good. But, you have to compete with already marketed solutions. Just take a look on university management software like https://www.cas-software.com/
So, your solution isn't required to do all the things the upper software does, but then you have to answer the question why your solution and not one of the established one's?
Is an integration with already used software possible? Because it's a sort of an act to change a running system.
Did you do a requirement analysis? What features, like ical export, can your solution cover?
There's a lot of pain in one's asstrich, that can be solved with disruptive, untraditional approach like yours. But one needs to know what and who's pains are relieved. How they're relieved. And what else can be done.
Just my 2 cents..
Also, I don't think it's easy like this. Teachers need's need more attention :) also, just click & assign doesn't solve the whole problem with assignments. A good teacher take care of the pupils and know what each of them are capable of. It's also the duty of a teacher to pull the pupils, not to push them.
This next bit is going to sound a bit harsh or confrontational, but trust I'm coming at it with a good heart of hope. I'm a bit on edge about pushing AI for grading. Teachers are... not technically literate folk. They won't understand the nuance of AI and using it to grade student work seems like heartache and disaster in the making. Are you doing anything to somehow subvert the fundamental nature of LLMs to make it suitable for grading?
reply