* Its absurdly good at coding and following types. For example if you change a type in Rust to be an Option, it will refactor the code to properly handle Options in the parts it used them. but it isn't perfect.
* It gets it well enough there. It can generate test cases so its easy to test to see if it works.
but in the end after hours and hours of trying to coax the AI, it was unable to do what I wanted, build a b-tree in Python. it built a binary tree just fine, but trying to have it generalize to a b-tree was a problem. * it introduced many many many subtle errors. Like variables not being initialized or not splitting the children correctly.
* Its implementation worked when all keys inserted were in order, but not when they were out of order.
* it would miss and leave out variables.
* It would frequently have index errors from trying to access lists out of bounds.
* writing the code in Rust was almost impossible. it would constantly have wrong types or move errors.
Overall I couldn't recommend this to anyone without a strong CS background. It introduces far too many subtle bugs in code that are are almost impossible to review because the code it produces is so convincing that you go "hmm maybe it knows what it is talking about" but in the end you have no idea what you should trust.Even the test cases that it generates can be deceptive. They look convincing but upon closer inspection, sometimes are just not really testing anything.