- USD$
- GBP£
- EUR€
- BRLR$
EN
Hands-on challenges in the main languages, auto-graded by tests with code quality analysis.
A coding assessment is a hiring test where candidates write code to solve a problem and the solution is run and graded automatically. In Coodesh coding challenges, the Coding question runs in an in-browser IDE and the score comes back immediately from unit tests that execute the candidate's code.
A resume and a chat tell you little about how someone actually codes. A hands-on challenge with test cases shows whether the code works, edge cases included, and applies the same bar to every candidate. When you care about style, structure, a full project or a specific framework, the same assessment can mix in other code formats.
Filter the library by level, skill and career track, or write your own Coding question with a prompt, typed inputs and outputs, and a reference solution.
Leave the question open to any language or restrict it to your stack. The test cases you add are the grading rule.
They write code in the browser, run the tests as often as they need and submit. Coding questions are taken on a computer, not on mobile.
Scores come per test case. The report shows performance by skill and integrity events, and the hiring decision always stays with your team.
Choose the format by what you want to measure: correctness, style, a project or a framework.
Candidates implement a solution with typed inputs and outputs, and automated tests grade it in real time.
Assess style and structure without strict tests. AI grades against criteria you write, and human review is always available.
Candidates deliver a full repository from a template on GitHub or in a .zip, and you read the code inside the platform.
A full in-browser environment for React, Vue, Angular, Django, Laravel, .NET and Spring Boot, with an AI policy you control.
Candidates record themselves explaining the solution. The platform adds 10 minutes to the question, and you can change the duration.
Coding answers show how the code was written, and every test run appears on the assessment timeline.
Test-based auto-grading lets you screen many candidates on the same criteria and bring only real problem solvers to interviews.
Limit languages to your stack and calibrate difficulty with Beginner to Advanced questions.
Write the challenge with your reference solution and test cases, take it yourself before sending it, and clone it for other roles.
Start for free, pick a challenge from the library or write your own, and invite your first candidates today.
The Coding question runs on Coodesh's own execution servers, which support dozens of languages. C, C++, Rust, COBOL and Dart were added in 2025. You can leave a question open so candidates pick their language, or restrict it to the languages your team uses. Frameworks are not part of this format: for React, Angular, Django, Spring Boot and similar, use the remote IDE question, which runs the full project in the browser.
The question author defines the solution name, typed inputs and outputs, a reference solution and the test cases. The candidate's code runs against those cases and the score is ready right away. Free coding and take-home projects are graded by AI against written criteria weighted 1 to 5. In every format a human reviewer can step in, and the hiring decision is always made by your team, not by the score alone.
Coding checks whether the code solves the problem, with unit tests and an instant score. Free coding focuses on style and structure, with no required tests, and is graded on criteria. Project is the take-home: the candidate delivers a full repository from a template. Remote IDE is a complete in-browser environment, with a terminal and a running app, for framework work and, if you allow it, AI use. You can combine formats in weighted sections.
Turn on the assessment's integrity features. They log pastes from external sources, text inserted without pasting, such as an AI assistant's insert button or a browser extension, tab switches, window focus loss and open developer tools. Coding answers also show how the code was written. These signals guide the review and never reject anyone on their own: your team decides what each event means.
Yes. Create a Coding question in your library, write the prompt and set language, time, level and weight. Then add inputs, outputs, your reference solution and test cases, and check everything in the review tab. Fill in the question's goal and how to assess it to guide reviewers. The question lives in your company's private library, and you can take it yourself before any candidate does.
Our ecosystem of deep integrations makes it easy to streamline your hiring processes.
Create your account and send your first assessment today.
