Our tool takes your practice test results page as an HTML input from your account and generates a comprehensive analysis, breaking down performance by skills, domain, and difficulty.
You'll even find out which adaptive module you got.
Plane.so is indeed a great tool, and they emphasize creating an unopinionated platform that can be utilized by anyone.
Our approach, however, is AI-first. We are focused on integrating multiple AI agents into our tool to assist stakeholders in various ways, including automating manual tasks, providing enhanced context, and even resolving issues end-to-end. Our goal is to leverage AI to significantly boost productivity and streamline workflows for our users.
Discover how we decoded the structure of four official tests using the College Board's Educator Question Library and even constructed new adaptive tests!
As the SAT transitions to a digital format in the US today, educators now have the opportunity to leverage the College Board's Question Library database. This enables the creation of custom, adaptive mock tests using official questions for a more tailored test preparation experience.
Wave looks really nice, though it’s a shame they only support US tax. For Australians the accounting package has to support “STP2” which is the governments API for reporting payroll and super, otherwise it’s borderline illegal to use.
Hey HN, we've been working on an interesting experiment at EdisonOS that might pique your curiosity. We're in the edtech platform space, and our primary offering is an infrastructure that allows anyone to set up their own online education platform without a dev team.
But here's the kicker: we've recently been playing around with ChatGPT to generate Digital SAT questions. As you may know, the College Board announced that the 96 year test is going Digital (already live in the international market) and only provides a limited set of four official mock tests for Digital SAT prep. We saw this as an opportunity to experiment with Gen AIs to create additional, practice material. We used ChatGPT to generate questions based on the format and style of the official SAT mock tests. The results were surprisingly good. Not only did the questions resemble the actual SAT questions in complexity and structure, but they also allowed us to create an adaptive testing environment.
We've documented our experience in a blog post, including a link to the ChatGPT-generated mock test if you're interested in taking the test yourself.
I would love to hear your thoughts on the potential (and pitfalls) of using machine learning to augment Testprep.
Have you also tried the document (containing questions and corresponding answer) uploading (for GPT4 beta) approach as this might save some time in writing very long prompts?
At TeachEdison, We rely on Heroku for all our backend services and been using for the last 24 months. Today, we received the following email about termination of services to Indian Customers who are using Credit Cards issued by Indian Banks.
"Effective December 15, 2021, Heroku is no longer accepting credit cards issued from India banks. Customers who are unable to utilize a different credit card will have until end of January 2022 to download their data, at which point their account will be suspended and then deleted. Enterprise India-based customers are not impacted. All paid services consumed prior to this time will be credited and existing credit cards will be removed from the account."
These six tests released by the College Board are the only official resources that closely resemble the new Digital SAT.
However, their reporting is very basic and doesn't help much in planning an effective prep strategy.
We've built a reporting tool to solve this problem. Sample Report: https://bluebook.edisonos.com/report/w1oqu29maxzwv40
Our tool takes your practice test results page as an HTML input from your account and generates a comprehensive analysis, breaking down performance by skills, domain, and difficulty.
You'll even find out which adaptive module you got.