Dimension scores are derived from public data and fields; weighted into the composite. Reference only.
OK is an assignment autograding and instructional feedback platform for computer science and data science courses. It covers student submissions, autograding, code review, personalized feedback, and course analytics. The text mentions that it has been used in courses such as CS 61A, Data Science 8, and CS 168, and has handled millions of submissions, making it clearly geared toward large-scale university teaching scenarios.
At the core of OK is a language-agnostic autograder: instructors only need to specify the commands to run, while the platform handles the grading infrastructure. It also allows instructors to view student solutions and leave comments on style improvements, reducing the need for paper-based grading workflows. Real-time statistics help instructors track assignment progress, the number of students who have completed work, and those who have not yet started. The feature list also includes plagiarism detection, online solution search, late submissions, group assignments, drag-and-drop uploads, Jupyter Notebook submissions, an assignment library, and Canvas LMS integration. These integrations are especially useful for data science courses and schools already using Canvas.
OK’s source code is available on GitHub, and instances can be run on your own servers. This is important for schools that want control over student code, grading results, and course data. The website provides documentation links, a GitHub repository, Issues, several video demos, and related educational research papers, indicating that it has accumulated a meaningful body of material around real teaching practice. However, the crawled text does not show deployment guides, APIs, SDKs, or operational requirements, so the completeness and update frequency of the documentation still need further verification.
For pricing, the hosted service is free if you teach a computer science or data science course at an accredited school. Policies for other organizations, commercial training providers, or non-accredited schools are not specified. The privacy policy states that the OK client sends code and autograder results to the server and stores information requested by instructors. The platform promises not to use personal information beyond running courses, servers, and instructional improvements, though it may use aggregated statistics in research publications. For institutions with strict compliance requirements, self-hosting may be more appropriate.
Its strengths are that it is open source, has clear free-hosting conditions, supports large classes, offers flexible grading methods, and provides a complete feedback and analytics loop. Its drawbacks are the lack of publicly available information on enterprise-level support, detailed deployment costs, payment methods, and pricing for non-university users. It is best suited for university CS/DS instructors, teaching assistant teams, and courses that need to manage programming assignments at scale.
Access from China is not covered in the text. The availability of GitHub, the hosted site, and video resources may need to be tested in practice. Payment methods are also not disclosed. If access or compliance is a concern, alternatives such as Gradescope, Autolab, Moodle/VPL, PrairieLearn, and CodeGrade may be worth evaluating.
⚠ This review is compiled from public sources and does not constitute a purchase recommendation. Verify all facts on the vendor's official site. Verify on okpy.org official site.
okpy.org is an United States Dev Tools provider. TG4G tracks its product information, an overall rating of 7.0/10, and a China-accessibility score of Workable. Click "Visit Official Site" to reach okpy.org directly.