Reusable role templates for technical screens: six kinds of platform, compared
Six kinds of platform give recruiters reusable role templates for technical screens: applicant tracking systems with interview kits, coding assessments, interview-as-a-service, interview intelligence, structured-interview tools, and AI or one-way video interviews. What separates them is who runs the screen — your recruiter, an outside engineer, software or nobody live — and who judges the answer.
Updated . Written by Hirezen, which makes one of the tools compared here — where it fits and where it doesn’t.
The six kinds at a glance
Every one of them lets you reuse a template across candidates for the same role. What separates them is who runs the screen and who judges the answer. The lines are blurring: some assessment platforms now sell AI interviewers, and some applicant tracking systems sell notetakers.
| Kind of platform | The reusable template | Who runs the screen | Who judges the answer | Examples |
|---|---|---|---|---|
| Applicant tracking systems with interview kits | An interview kit on each stage of a job: the questions to ask and a scorecard to fill in | Your own interviewer | Your interviewer, on the scorecard | Greenhouse, Ashby, Lever, Workable, Pinpoint |
| Coding assessment platforms | A test for the role, built from a library of coding tasks and skills questions | Nobody live: the candidate takes it alone, or an engineer runs a live coding pad | The platform scores it; an engineer reviews | HackerRank, CodeSignal, Codility, CoderPad, TestGorilla |
| Interview-as-a-service | A structured interview and scoring rubric for the role, run by the vendor’s interviewers | An engineer the vendor supplies | The vendor’s interviewer, in a scorecard or report | Karat, BarRaiser, TechScreen |
| Interview intelligence tools | Note and scorecard templates the software fills from a recording of the call | Your own interviewer, with the software recording the call | Your interviewer, with a transcript and AI notes to check against | Metaview, BrightHire, Greenhouse Notetaker |
| Structured-interview tools | A plan for each round: the questions, what each is for, and the signals to listen for | Your own interviewer: a recruiter on the first screen, an engineer after it | Your interviewer marks what they heard; the team decides | Hirezen, Glider AI |
| AI and one-way video interviews | A question list per role, answered on video, or a role template an AI interviewer follows | Nobody live: recorded questions, or software that asks follow-ups | Your team reviews the answers later, or the software scores them | HireVue, HackerRank Chakra, CodeSignal AI Interviewer, Fabric, Alex (formerly Apriora) |
Kind 1
Applicant tracking systems with interview kits
Many applicant tracking systems let you attach an interview kit to each stage of a job — the questions to ask and a scorecard to fill in — and reuse it on the next opening. The ATS supplies the container; someone on your team still has to write what a strong answer sounds like.
- Fits when
- You already run hiring in an ATS, and an engineer on the team will write and maintain the kit for each role.
- What a new recruiter still needs
- Content. A kit is a form to fill in; a new recruiter needs each question’s purpose and the signs of a strong answer written into it, or they record words they cannot judge.
Examples
- Greenhouse — an interview kit for each interview — focus attributes, questions and interviewer prep — whose focus attributes head the interviewer’s scorecard.
- Ashby — shared interview plans, which work as templates across jobs.
- Lever — structured interview kits, prescriptive feedback forms and customizable scorecards.
- Workable — interview kit templates — six by default, or imported from another job — which interviewers fill in as a scorecard.
- Pinpoint — candidate scorecards from reusable company templates, with weightings that give a criterion more influence on the candidate’s overall result.
Kind 2
Coding assessment platforms
Assessment platforms keep a library of coding tasks and skills tests and let you save a test per role. The candidate takes it alone, and the platform scores it automatically, so a recruiter can screen without judging code — they read a score and decide where the cut-off sits.
- Fits when
- You screen many candidates for the same role, or no engineer has time for first-round screens.
- What a new recruiter still needs
- A cut-off they can defend, and a conversation. A score says whether the code passed, not how the person works, and an unsupervised test now has to account for AI help.
Examples
- HackerRank — tests built from a predefined role, with its mapped skills and curated questions, or generated from a job description.
- CodeSignal — Certified Assessments built by subject-matter experts and validated by industrial-organizational psychologists.
- Codility — ready-made Screen templates for common tech roles at two or three seniority levels, edited to fit the job.
- CoderPad — Screen tests assembled from a role, an experience level and the skills to test.
- TestGorilla — a test library with programming and role-specific tests.
Kind 3
Interview-as-a-service
Interview-as-a-service firms, also sold as technical screening services, supply the engineer who runs your technical screen, working from a structured interview and a scoring rubric for the role. Your recruiter books the interview and reads the scored result or report; nobody on your team has to judge the answer.
- Fits when
- Engineering time is the bottleneck, and the volume justifies paying per interview.
- What a new recruiter still needs
- A check that the vendor’s bar matches your team’s before trusting its reports, and a plan for the candidate, whose first technical conversation is with someone outside your company.
Examples
- Karat — live interviews run by certified Interview Engineers against consistent scoring rubrics.
- BarRaiser — expert interviewers who run structured rounds around the clock, with a scorecard ending in a hire / no-hire recommendation.
- TechScreen — Screening-as-a-Service: role-specific interviews built from the role’s tech stack and deliverables and validated with your hiring managers, run with at least an hour’s notice, with its Verinalysis report back within minutes of the interview’s end.
Kind 4
Interview intelligence tools
Interview intelligence tools record and transcribe your own interviewer’s call, then draft notes against a template or the scorecard in your ATS. For a new recruiter the recording is the safety net: the hiring manager can replay a technical answer instead of trusting the recruiter’s notes of it.
- Fits when
- Your candidates accept a recorded interview, and the hiring manager will actually review the recordings.
- What a new recruiter still needs
- The questions and what a strong answer sounds like — a recording captures whatever was asked — and consent to record, which some US states require from everyone on the call.
Examples
- Metaview — transcribes the call and fills in note templates built from your scorecard, intake questions or debrief framework.
- BrightHire — AI-drafted structured interview plans with competencies, stages, skills-based questions and evaluation rubrics.
- Greenhouse Notetaker — joins Zoom, Meet or Teams interviews and, with the candidate’s consent, records and transcribes them; notes drafted for each question go onto the scorecard in one click.
Kind 5
Structured-interview tools
Structured-interview tools give your own interviewer a plan for each round: the questions, what each one is for, and the signals of a strong answer. A recruiter can run the first screen from it, marking what they hear, and hand the engineer’s round a record of which signals came up and which did not.
- Fits when
- Your own people run the screens, and each of them interviews differently today.
- What a new recruiter still needs
- A clear hand-off. The recruiter screen checks experience, scope and communication against the signals; depth of skill is for the engineer’s round after it.
Examples
Kind 6
AI and one-way video interviews
The candidate answers on their own. One-way video platforms record answers to a set question list for your team to review later; AI interviewers ask the questions themselves, follow up and score the answers against a rubric. Either way screening stops waiting on anyone’s calendar, and recruiters review instead of running calls.
- Fits when
- You screen at a volume no team could interview live, and your candidates accept answering to a camera or to software.
- What a new recruiter still needs
- A rubric for reviewing answers, and the rules where you hire: New York City requires a bias audit and candidate notice for automated hiring tools, Illinois requires notice and consent before AI analyzes a recorded video interview, and the EU AI Act treats AI that evaluates candidates as high-risk.
Examples
- HireVue — on-demand video interviews, where candidates record answers to a list of questions on their own schedule, alongside live ones.
- HackerRank Chakra — an autonomous AI interviewer with an in-line code editor, scoring every candidate against the same rubric.
- CodeSignal AI Interviewer — AI interview agents that run a short interview, typically 15 minutes, and return a scored report with the full transcript.
- Fabric — AI-run live coding interviews that execute the candidate’s code, ask follow-ups about complexity and edge cases, and score correctness, code quality, debugging and communication.
- Alex (formerly Apriora) — AI interviews by video, phone or chat, customized to the role, with role-fit scores and searchable transcripts.
What makes a technical screen credible when a recruiter runs it
A screen is credible when two recruiters would make the same call on the same candidate, and the engineers who meet that candidate next would agree with it. Structure is what gets you there: in the largest recent re-analysis of hiring research, structured interviews predicted job performance better than any other common selection method.
Every candidate for the role gets the same questions
Otherwise the difficulty of the question becomes the difference between candidates. A reusable template is how that survives a new recruiter, and it is the “structured” in structured interview: in a 2022 re-analysis, structured interviews had the highest average validity of the common selection methods (.42), ahead of job knowledge tests (.40) and work samples (.33). Sackett, Zhang, Berry & Lievens, 2022
Each question says what it is for
One line of purpose — “checks they have run a system in production, not only built one” — tells a recruiter which part of the answer matters and when to ask the follow-up.
The template says what a strong answer contains
Signals in plain words, such as “names a trade-off they made” or “gives numbers for the scale they worked at”, are what someone who is not an engineer can hear and mark. Without them the recruiter writes down words they cannot judge.
When an answer is outside your expertise, ask the candidate to explain it
That is the advice Triplebyte’s founders gave at Y Combinator’s Startup School for an interviewer outside their own field. Bartram & Taggar, 2018, at 49:55
The screen decides who gets an engineer’s time; it does not grade code
A recruiter screen checks experience, scope and communication against the signals. Depth of skill is for the next round — an engineer’s interview or an assessment — and the screen’s notes tell that round what to probe.
A 30-minute technical screen a recruiter can run
A template for screening software engineers, written so a recruiter who is not an engineer can run it. Four questions score three competencies, and each question comes with what it is for, the signals of a strong answer in plain words, and a follow-up. It decides who gets an engineer’s time; it does not grade code.
Technical screen · Software engineer · 30 min
- What it scores
- Hands-on experience: what they built and ran themselves, as distinct from what their team did.
- Communication: whether they can make a technical decision clear to someone who is not an engineer.
- Scope: how much they have owned — a feature, a service, a whole system — against the level you are hiring for.
- Before the first call
- Ask the hiring manager what a candidate at this level should have owned — a feature, a service or a whole system, and for how many users — and what earns the next round. The scope question is marked against the first, and your decision after the call against the second. Settle salary, notice period and location beforehand, in writing or on another call, so this one is spent on the questions.
- While you run it
- Ask every candidate for the role these questions, in this order, and move on when a question’s time is up, so that everyone is asked all of them.
- Mark each signal yes, no or not sure as you hear it, on what they said rather than how they said it: accent, fluency and confidence are not signals.
- When an answer goes past what you know, ask them to explain it in plain words. It is a fair question, and it gets you an answer you can mark.
Opening · 2 min
Say who you are, how long the call is, and that it is about work they have done. Tell them the next round is with an engineer on the team, who goes deeper, so there is no coding today, and that you will take notes and may ask them to explain technical parts in plain words.
Question 1 · 7 min
Walk me through something you’ve built that you’re proud of. What did it do, and what was your part in it?
Scores: Hands-on experience
Purpose: Separates what they did from what their team did. If they only say “we”, ask which part was theirs.
Listen for
- Says which parts were theirs and which were the team’s
- Says what it did for the people who used it, not only which technologies it used
- Gives the reason behind a decision they made
- Describes a problem they ran into along the way
Follow-up: What would you change if you built it again?
Question 2 · 6 min
What’s the hardest bug or outage you’ve worked on? How did you find the cause?
Scores: Hands-on experience
Purpose: Checks they have run a system in production, not only built one; for a junior role, a bug in a project or an internship counts. If the story jumps from the problem to the fix, ask how they found the cause.
Listen for
- Says what they suspected, and how they checked it themselves
- Names what pointed to the cause — an error message, a log, a dashboard, a failing test
- Says what they changed so it would not happen again
Follow-up: How did you know the fix worked?
Question 3 · 6 min
Tell me about a technical trade-off you made, the way you’d explain it to a product manager who had to agree to it. What were the options, and what did you give up?
Scores: Communication
Purpose: Checks they can make a technical decision clear to someone who is not an engineer — for these few minutes, you. Ask about any word you don’t know.
Listen for
- Describes the options without jargon, or explains each term you ask about
- Gives a reason tied to users, deadlines, cost or the team, not only that one option is better
- Names a real cost of the option they chose, not only its benefits
- Uses an example or a comparison to make it concrete
Follow-up: What would have made you choose the other option?
Question 4 · 5 min
Think of the system you know best. How many people used it, roughly how much data or traffic did it handle, and how many engineers worked on it?
Scores: Scope
Purpose: Checks how much they have owned against the level you are hiring for. If a number is confidential, an order of magnitude will do.
Listen for
- Gives numbers, even rough ones
- Keeps the size of their own part apart from the size of the company
- What they owned fits what the hiring manager said the level needs
Follow-up: What was the hardest part of running it at that size?
Close · 4 min
Ask what questions they have, and answer what you can. A technical question you can’t answer is for the engineer’s round: say so, and pass it on. Then tell them what happens next and when they will hear from you.
After the call
Decide against what the hiring manager said earns the next round. Your marks are the record of which signals came up and which didn’t. If the candidate goes through, send the engineer who runs the next round the gaps — every signal marked no or not sure, and any you never reached — so that round starts there. The engineer goes deeper with questions of their own rather than asking yours again, since the same question only gets the same prepared answer.
To run it in Hirezen, name the three competencies on the position, add a 30-minute round, give its opening 2 minutes and its close 4, and write these questions into it, each with its purpose, signals, follow-up and the competency it is for. Whoever runs the screen then marks each signal yes, no or not sure as they hear it.
Which kind to choose
Start from who is available to run the screen, not from the feature list.
No engineer can spare time for first screens
A coding assessment, where the candidate works alone and the platform scores it, or interview-as-a-service, where another company’s engineer runs it.
Your recruiters run the first screen
A structured-interview tool or a well-written ATS kit, as long as every question carries its purpose and the signals of a strong answer, with an engineer’s round after it.
The hiring manager wants to check the recruiter’s read
Interview intelligence, if candidates consent to a recording the hiring manager will replay; or a plan whose marks show which signals came up, if that record is enough.
You screen hundreds of candidates a week
An assessment, one-way video or an AI interviewer, if your candidates accept one — and check the rules on automated hiring tools where you hire before you switch it on.
Where Hirezen fits, and where it doesn’t
Hirezen is one of the structured-interview tools above. Its library has 206 free interview plans for 39 roles, one per round — technical interviews, coding rounds, system design and more — and every question comes with what it is for and the signals to listen for. A team starts a position from a plan or writes its own, keeps the questions it trusts in a shared library, and gives every interview the same opening and close.
In the interview, whoever runs it — a recruiter on the first screen, an engineer on the technical rounds — marks each signal yes, no or not sure as they hear it. AI drafts the write-up from those marks, and the debrief puts every interviewer’s read side by side.
It does not set coding tests or grade code, it does not supply interviewers, and no bot joins the call: nothing is recorded. If you want an automated coding test, pair it with one of the assessment platforms above.
Questions
- Can a recruiter who is not an engineer run a technical screen?
- Yes, if the template does the technical part: the same questions for every candidate, a line on what each question is for, and the signals of a strong answer in plain words. The recruiter marks what they hear and asks the candidate to explain anything outside their expertise. Judging code itself is for an engineer or an assessment in the next round.
- What should a reusable technical screen template include?
- The questions in order with a time for each, what each question is for, the signals of a strong answer, follow-ups for when an answer is thin, and how the interview opens and closes. Hirezen’s interview plans carry all of these, and the page on how plans are structured shows what each part is for.
- Are free interview templates good enough for technical screens?
- A free template is a starting point, not a finished screen. Check that each question has a purpose and signals, cut what your role does not need, and make sure every interviewer for the role uses the same version. Hirezen’s plans are free to read without signing up, and using one opens it in a free workspace.
- Do I need an applicant tracking system to use interview kits?
- For an ATS’s own kits, yes: they live on the jobs in that ATS. Structured-interview tools and assessment platforms work without one. Hirezen keeps positions, candidates and interviews in its own workspace, and a write-up copies out formatted for a Greenhouse scorecard or as plain text.
- Should an AI interviewer run first-round technical screens?
- It can, at volume, if your candidates accept it and you meet the rules where you hire: New York City requires a bias audit and notice for automated hiring tools, Illinois requires consent before AI analyzes a recorded video interview, and the EU AI Act treats AI that evaluates candidates as high-risk. Hirezen does not interview anyone itself: no bot joins the call, and people make the decision.
- Which platforms have customizable scoring templates with recruiter-editable weightings?
- Some applicant tracking systems and assessment platforms do; Hirezen does not. Pinpoint’s candidate scorecards take weightings that give a criterion more influence on the candidate’s overall result, and some assessment platforms let you weight each test in an assessment so a core skill counts for more. Hirezen does not weight competencies and has no weightings to edit: whoever runs the interview marks each signal yes, no or not sure, AI drafts the write-up from those marks, and the debrief puts every interviewer’s read side by side.
- Which interview platforms have a structured, competency-based question bank for technical roles?
- Structured-interview tools do; an applicant tracking system’s interview kit holds the questions your own team writes, and assessment platforms keep a bank of coding tasks and tests instead. Hirezen’s bank is its plan library: one plan per round, every question with what it is for, the signals of a strong answer and a follow-up, and a question can be written for one of the position’s competencies. The questions a team trusts stay in a shared library in its workspace. The library covers engineering and IT roles — software, cloud and infrastructure, security and compliance, machine learning — with a handful for search and marketing roles, and the Hirezen section of this page counts its plans and roles as they stand. It has no plans for data science or product roles.
- Which platforms provide prebuilt role criteria for engineering, data science and product roles?
- Assessment platforms reach furthest across the three: Codility’s ready-made Screen templates include a Data Scientist role beside the developer ones, and TestGorilla’s test library has role-specific tests for data science and product management as well as programming. In Hirezen, role criteria are the competencies a position names, each flagged must-have or not, and the signals under every question, which are what the interviewer marks. The prebuilt ones are the library’s plans for engineering and IT roles, each with its key skills, questions and signals; a team hiring for data science or product writes its own competencies and questions or adapts the nearest plan.
Sources
Claims about a product come from its maker’s own pages, read on October 12, 2026. Products change; if one of these has, tell us at hi@hirezen.io.
- Sackett, Zhang, Berry & Lievens (2022), “Revisiting meta-analytic estimates of validity in personnel selection”, Journal of Applied Psychology
- SIOP, “Is cognitive ability the best predictor of job performance?” — the 2022 estimates by method
- Ammon Bartram & Harj Taggar, Y Combinator Startup School (2018)
- NYC Department of Consumer and Worker Protection, Automated Employment Decision Tools
- Illinois Artificial Intelligence Video Interview Act (820 ILCS 42)
- Regulation (EU) 2024/1689, the AI Act — Annex III, point 4