FAQ
Questions, answered
GetGud is free while in beta. If something here is unclear or missing, tell us on GitHub.
Pricing and the beta
How much does GetGud cost?
Nothing while we are in beta. Every feature is available, and there is no card to enter.
What happens when the beta ends?
We’ll publish pricing before the beta ends, before any paid plans take effect. The interviews, reports and questions you created stay yours either way.
What does “beta” mean here?
The core flow works end to end: post a job, plan an interview from an application, run it live and read the evidence. Expect rough edges and frequent changes. Your feedback decides a lot of what changes next.
Feedback and support
Where do I report a bug or request a feature?
Open an issue on GitHub. You can also see what others have reported and add to it. That repository holds feedback only; GetGud's code is not public.
Issues are public, so never include candidate names, résumés or interview answers.
How do I reach you privately?
Email tim@adrenal.ai for anything involving candidate data, your account, or a security concern.
The product
What does GetGud do?
It turns a job's requirements and an applicant's résumé into a focused engineering interview, with a reason for every question. You run the interview live, record a judgment on each answer, and get a report of what the interview established for each competency. See a sample interview.
Is GetGud an applicant tracking system?
No. We cover the path from an application to interview evidence: your jobs page, applications, interview plans, live interviews and reports. Keep using your ATS for the rest of hiring.
Does GetGud score or rank candidates?
No. There is no overall score, ranking or hire recommendation. Interviewers judge each answer as weak, mixed, strong or exceptional, and the report shows that evidence per competency with the notes behind it. Objective answers are checked automatically, but accuracy is never turned into a hiring score.
Do résumé claims count as evidence?
Never. Claims decide what the interview tests. Only what the candidate shows in the interview counts, and the report says when a claim was not supported.
Where do the questions come from?
We write and review them ourselves. Each is built on a problem engineers hit at work, such as a race condition, a misleading query plan or a prompt injection, and comes with an interviewer guide and rubric. You can also add your own questions to your library.
Which roles does it cover?
Frontend, backend, full stack, platform, SRE, security and applied AI engineering, at levels from junior to staff. Coverage is deepest for TypeScript, React and Python, and still growing for Java, Go, DevOps and ML.
Do I need a job post or a résumé to use it?
No. You can build an interview from a role template or pick questions yourself. A job post and an application let us tailor the plan to the role and the candidate.
Interviews and candidates
What is the interview like for the candidate?
They open a link, enter their name and answer questions: multiple choice, select all that apply, ordering steps, or choosing lines of code. No account is needed. You watch their progress live and discuss each answer with them.
Can candidates use AI tools?
You decide per interview: prohibited, allowed or encouraged. We do not watch the candidate's screen, camera or microphone. We record activity counts, such as how often the window lost focus and how many paste events happened, never what was pasted.
Accounts and data
How do I sign in?
With a Google or Microsoft account, ideally your work account. There is no GetGud password.
Who can see candidate data?
Only members of the workspace that created the interview. When a workspace deletes an interview, we delete it within 30 days. See the privacy policy.
Which service providers do you use?
Supabase (database, authentication and realtime messaging), Vercel (application hosting), Google Cloud (file storage for company logos and applicant resumes), Google and Microsoft (interviewer sign-in). We do not sell personal data.
Found a bug or missing something?
Open an issue. We review feedback submitted there.