- USD$
- GBP£
- EUR€
- BRLR$
EN
Open-ended questions that measure clarity, argumentation and subject knowledge, graded with a rubric and comparable across candidates.
I would start by mapping each request to the quarter’s business goals, so the discussion moves from opinions to impact. Then I’d estimate effort and risk with the team and share a ranked list openly. When two requests still conflict, I bring both stakeholders together with the data and agree on a trade-off we can revisit next sprint.
A written assessment for hiring is a test in which candidates answer open-ended questions in their own words, showing how they structure ideas, make an argument and handle the subject. On Coodesh, you build it with Free text questions graded against criteria your company defines: the AI applies the same rubric to every answer, and your team reviews before any decision.
Multiple choice shows whether someone recognizes the right answer. An essay question shows whether they can build one, explain their reasoning and write clearly, which matters in almost any role that communicates in writing. Grading time used to be the blocker. With written criteria and AI grading, open-ended questions work even when you have many applicants.
Go to Assessments › Library › Create question, choose Free text, and write the title and question in the job's language.
Set the answer time, the level (beginner, practitioner or advanced), the weight and the skills the question covers.
Describe the goal of the question and how to grade it, then add criteria such as clarity, argument and subject knowledge, each with a weight.
Once the candidate submits, the AI grades against your criteria and the score updates. Your team reads the answers and decides who moves on.
Grading follows written criteria, and integrity tracking records signals for the reviewer.
Each criterion has a name, a description and a weight from 1 to 5, so what matters most for the role counts most.
Mark a criterion as extra to leave a note for the reviewer without changing the score.
Grade with AI is on by default. Switch it off on the question, or in one assessment only, and answers go to manual review with the criteria as a guide.
When enabled, pasted text is stored, and pastes that did not come from the same question weigh more in the integrity indicator.
The platform logs text that enters an answer without going through the clipboard, such as through an AI assistant's insert button.
Add files in the question editor, such as a source text or a chart for the candidate to analyze.
Support, customer success, product and leadership roles depend on clear messages. A written answer shows that before you hire.
A question like how to prioritize a backlog with conflicting requests shows how a person thinks, not just what they know.
An open answer in English or Spanish shows real writing in the job's language, alongside the language test.
Create a free Coodesh account and write your first open-ended question with a grading rubric.
The AI reads the answer and gives each criterion on the question a score from 0 to 100%. The question score is the weighted average of those criteria, using the weights you set. While the AI grades, the question shows as Processing in the result. If it cannot grade an answer, the question goes to manual review. Criteria marked as extra are not counted in the score.
Yes. Switch off Grade with AI on the library question or only for one assessment, without changing the original question. Answers then go to manual review, and the criteria stay visible to guide whoever grades. Questions with no grading criteria also go to manual review. While any question is pending, the assessment score is shown as partial.
The platform does not claim to know who wrote a text, but with integrity features on it logs signals: pastes, with the pasted content stored, text inserted without pasting, for example through an AI assistant button or a browser extension, leaving the tab and the window losing focus. These are signals for the reviewer, not proof of cheating. Your team reviews them and decides.
Yes. Once graded, the question score feeds the section score, and the overall assessment score is a weighted average of the sections based on the weight of each one. That lets you give writing more weight when it is central to the role. Results also show the score by the skills linked to each question.
No. The AI score puts every answer on the same rubric, but moving someone forward or ending their process is always a decision made by a person on your team, who reads the answer first. The score helps you prioritize the review; it does not replace it.
Our ecosystem of deep integrations makes it easy to streamline your hiring processes.
Create your account and send your first assessment today.
