Product
The technical part of hiring, on real code. Evidence, not verdicts.
Kendor runs coding screens, live pair coding and interviews in sandboxes that look like the job: real runtimes, real databases, your own repositories. Candidates work in a browser IDE. Your reviewers see what happened, each record their own verdict, and the decision stays with people.
Coding screens
Async take-homes in a real browser IDE, graded on a clean, version-pinned re-run with the network cut.
Read moreLive pair coding
One link and no account for the candidate. Editor, terminal, running app, whiteboard and call in one shared sandbox.
Read moreInterviews
Candidates book into your availability. Invites update in place, and you can review a submitted screen together.
Read moreAI agents
Claude Code and Codex in the candidate's sandbox, on your own keys, with token budgets you set. Off by default.
Read moreReal environments
Seven languages at pinned versions, PostgreSQL and Valkey in the sandbox, multi-service apps in one kendor.yaml.
Read moreRepo import
GitHub, GitLab, Bitbucket or your own Git server, private repos too. Each import is a fixed snapshot and can take one folder of a monorepo.
Read more
How it fits together
One environment, from first screen to final decision.
Build the screen from real work
Import a repository from GitHub, GitLab, Bitbucket or your own Git server, describe its services in kendor.yaml, and choose whether agents are allowed.
Candidates work in the browser
An async screen sent from a job posting or an Ashby stage, then a live pair coding session or an interview where you can open their submission together.
Reviewers decide on evidence
A clean re-run for the score, integrity signals for context, a verdict from each reviewer, and a person who advances or rejects.
Run your next screen on Kendor. Read what the code says.
Kendor works with engineering teams in the EU, the UK and the US. Tell us what you're hiring for and we'll set up your organization with you.