All features

Draft scores in a click. Graded on the candidate, not their AI.

Draft with AI reads the submission against your rubric and proposes a score and notes for every criterion. It also weighs how the candidate directed the assistant, so wholesale delegation does not score like real work.

Start freeSee pricing

Screenshot coming soon

Use cases

When this feature earns its keep.

Three concrete patterns. Use them as templates for your own setup.

A faster first pass

One click reads the submission against your rubric and fills in a proposed score and a note for each criterion. You edit, you override, you decide. It just removes the blank-page part.

Catches delegation, rewards direction

The grader reads the candidate AI transcript alongside the code. When the prompts show meaningful problem-solving it counts; when they show "do the whole thing for me", the scores reflect the candidate, not the assistant.

The same rubric, every time

Every submission is drafted against the same criteria with notes attached, so your scoring stays consistent across candidates and your reviewers calibrate to one standard.

Setup

How the setup looks.

AI grading lives in the scoring panel next to your rubric. It proposes; you dispose.

Same workspace, same auth, same collaboration features as every other Typelets workspace. See all features

  1. Define the rubric criteria

    Set the criteria you score against and an optional description for each. These are exactly what the draft scores against.

  2. Open a submission and Draft with AI

    From the scoring panel, pick the candidate and hit Draft with AI. It reads their submitted files, the rubric, and their AI transcript, then proposes scores and notes.

  3. Review, adjust, and apply

    The draft fills only unscored criteria and empty notes, so nothing you already typed is overwritten. Edit anything, then save. The final scores are yours.

FAQ

Does the AI decide who gets hired?
No. It drafts a first pass you are free to change or ignore. Applying a draft only fills criteria you have not scored and notes you have not written; your edits always win, and the decision is yours.
Does it penalize candidates for using AI?
No. Well-directed AI use is normal modern practice and is never a penalty. What moves the scores is delegation: if the transcript shows the candidate handed the whole task to the assistant, the submitted code reflects the assistant, not them, and the grading says so.
What does the grader actually read?
The submitted files, your rubric and criteria, and, when the assistant was enabled, the candidate AI transcript. It grades primarily on the code and uses the transcript to judge how much of it is the candidate's own work.

Start interviewing in under a minute.

No card, no install, no yaml. Sign up, create a workspace, share the link. That's it.

Free during beta · No credit card · Cancel anytime