# Job application form template that screens as it collects Short answer: this is a complete QuizGen form definition for a job application — name, contact, an uploaded resume, a portfolio link, a few screening questions, and a scoring block that sorts applicants into "screen now," "review," or "not a fit" as they submit. POST it as-is, or hand it to an AI assistant to adapt for your role. ## Resume uploads need a paid plan Candidates can attach a resume directly: the `file_upload` question type takes PDFs and Word documents, and the file lands in your account rather than in a share link that may or may not still work in three weeks. The catch is plan gating. `file_upload` is **Pro or Business only** — on the free plan or in the sandbox, creating a form that contains one returns 402 `plan_limit`. Limits differ by plan too: Pro allows 10 MB per file and 1 GB of storage in total, Business 25 MB per file and 10 GB. If you're on the free plan, delete the `resume_file` question from the template below and keep `portfolio_link` — a `url` question asking for a Drive link, personal site, LinkedIn, or GitHub. That's a genuinely fine substitute for engineering, design, and writing roles, where a live portfolio often tells you more than a PDF. It's a worse substitute when you need a formatted document to pass along to someone else. The template below asks for both, which is usually the right call: the file for your records, the link for the work itself. One thing worth knowing if an assistant is building this for you: the answer to a `file_upload` question is an array of file ids that the server issues while the candidate uploads. You never write that value into a definition, and there is no API for attaching a file on someone's behalf — uploads only happen through the hosted form. ## The template ```json { "schema_version": 1, "kind": "form", "title": "Apply: Product Designer", "description": "Takes about 5 minutes. We read every application.", "settings": { "collect_respondent": "email_required" }, "sections": [ { "key": "candidate", "title": "About you", "questions": [ { "id": "full_name", "type": "short_text", "prompt": "Full name", "required": true, "placeholder": "Jordan Smith" }, { "id": "email", "type": "email", "prompt": "Email address", "required": true, "placeholder": "you@example.com" }, { "id": "phone", "type": "phone", "prompt": "Phone number", "required": true, "placeholder": "+1 555 555 5555" }, { "id": "resume_file", "type": "file_upload", "prompt": "Upload your resume", "help": "PDF or Word document.", "required": true, "max_files": 1, "accept": ["pdf", "document"] }, { "id": "portfolio_link", "type": "url", "prompt": "Link to a portfolio, LinkedIn, or GitHub", "help": "Optional, but it usually helps more than the resume does.", "placeholder": "https://" } ] }, { "key": "role", "title": "The role", "questions": [ { "id": "position", "type": "choice", "prompt": "Which role are you applying for?", "required": true, "options": [ { "value": "product_designer", "label": "Product Designer" }, { "value": "senior_product_designer", "label": "Senior Product Designer" }, { "value": "design_intern", "label": "Design Intern" } ] }, { "id": "years_experience", "type": "number", "prompt": "Years of relevant experience", "required": true, "min": 0, "unit": "years" }, { "id": "authorization", "type": "choice", "prompt": "Work authorization and location", "required": true, "options": [ { "value": "us_remote", "label": "Authorized to work in the US, remote" }, { "value": "us_onsite", "label": "Authorized to work in the US, near our office" }, { "value": "sponsorship_needed", "label": "Would need visa sponsorship" }, { "value": "outside_us", "label": "Based outside the US" } ] }, { "id": "key_skills", "type": "multi_choice", "prompt": "Which of these do you have hands-on experience with?", "min": 1, "options": [ { "value": "figma", "label": "Figma" }, { "value": "design_systems", "label": "Design systems" }, { "value": "user_research", "label": "User research" }, { "value": "html_css", "label": "HTML/CSS" }, { "value": "prototyping", "label": "Prototyping (motion/interaction)" } ] }, { "id": "why_role", "type": "long_text", "prompt": "Why this role, and why us?", "rows": 5, "required": true, "examples": [ "I've used your product for two years and want to fix the onboarding flow I struggled with" ] } ] }, { "key": "consent", "title": "Before you submit", "questions": [ { "id": "data_consent", "type": "consent", "prompt": "I consent to this application, including my contact details and answers, being stored and reviewed by the hiring team for this role.", "required": true } ] } ], "outcomes": { "method": "points", "points": { "authorization": { "us_remote": 3, "us_onsite": 3, "sponsorship_needed": 1, "outside_us": 0 }, "key_skills": { "figma": 1, "design_systems": 2, "user_research": 1, "html_css": 1, "prototyping": 1 } }, "buckets": [ { "key": "not_a_fit", "max": 2, "title": "Thanks for applying", "body": "We'll keep your application on file and reach out if a better-matched role opens up." }, { "key": "review", "min": 3, "max": 5, "title": "Application received", "body": "Your application is in the queue for review — we'll follow up within two weeks either way." }, { "key": "screen_now", "min": 6, "title": "Thanks — you're a strong match on paper", "body": "We'll reach out to schedule a first call within the next few days." } ] } } ``` Note on `years_experience`: `outcomes.points` scores option values, which only exist on choice-shaped questions — a plain `number` field can't be scored directly. `years_experience` is collected here for display and manual review, not scoring. If you want years of experience to feed the score (e.g. 0–1 years = 0 points, 2–4 years = 2, 5+ years = 4), convert it to a `choice` question with those bands as options and score each option in `outcomes.points` the same way `authorization` is scored below. ## What each piece is doing - `resume_file` is a `file_upload` limited to `["pdf", "document"]` with `max_files: 1` — resumes only, so you don't get a folder of screenshots. Requires Pro or Business; drop this question on the free plan. - `portfolio_link` is a `url`, deliberately optional. Asking for both a file and a link costs the candidate nothing and covers both the "send this to the hiring manager" case and the "actually look at their work" case. - `authorization` as a `choice` (not free text) means you can score it and filter responses by it later — much easier than parsing sentences. - `key_skills` is `multi_choice` with `min: 1`; each selected skill adds its own points in `outcomes.points`, so someone who checks three relevant skills scores higher than someone who checks one. - `data_consent` is a required `consent` question — a timestamped record that the candidate agreed to have their application stored and reviewed, which is worth having regardless of your local privacy-law requirements. - `outcomes` with `points`/`buckets` is the differentiator here: instead of just collecting applications, QuizGen scores each one as it comes in and stores `score` and `outcome` on the response, so you can sort your dashboard by "screen_now" first instead of reading 200 applications in submission order. ## This is a triage aid, not a hiring decision Say this plainly to whoever's using it: automated scoring should narrow the queue, not make the call. A few rules worth following: - Score only job-relevant criteria — years of experience, required authorization, named skills. Never score name, address, alma mater, or anything that correlates with a protected characteristic. - Keep a human reviewing every bucket, including "not_a_fit" — automated screening is good at surfacing the strong matches faster, bad at being the final word on a candidate. - Check your local rules. Some jurisdictions (New York City's Local Law 144 is the well-known example) regulate automated employment decision tools specifically — know whether yours applies before you rely on scoring for more than internal triage. The same `outcomes` mechanism — points per answer, buckets with a title/body — is exactly what powers [lead qualification forms](/blog/lead-qualification-form-that-scores-itself) and general [quizzes with scoring](/blog/how-to-make-a-quiz-with-scoring); if you haven't used `outcomes` before, either of those is a good second read. ## Route new applications into Slack or your ATS Don't make anyone babysit a dashboard. Set a [webhook](/blog/quiz-response-webhooks) on the quiz and every completed application POSTs to your endpoint immediately, with the answers, `score`, and `outcome` bucket included — pipe it into a Slack channel, or into whatever ATS or spreadsheet already runs your hiring pipeline. ## Let an AI assistant fill it in Connect [QuizGen's MCP server](/mcp) and say something like: > Read https://quizgen.dev/blog/job-application-form-template.md and build > me an application form for a "Backend Engineer" role — require 3+ years > of experience, score Python and Postgres skills highest, and require US > work authorization to land in the screen_now bucket. Give me the link. ## FAQ **Can candidates actually attach a PDF resume?** Yes, on Pro or Business — that's the `file_upload` question. On the free plan or in the sandbox it returns 402 `plan_limit`, so drop the question and rely on the link. **How big a file can they send?** 10 MB per file on Pro, 25 MB on Business, against 1 GB and 10 GB of account storage respectively. A resume will never come close; a reel or a design portfolio might. **Can I accept a cover letter too?** Raise `max_files` (up to 5) on the same question, or add a second `file_upload` for cover letters specifically. Keep `accept` narrow either way. **Can I require the resume but not phone?** Yes — toggle `required` per question independently; nothing forces you to require every field. **Does the scoring reject anyone automatically?** No. `outcomes` only sets a `score` and `outcome` bucket on the stored response and (optionally) a different completion message — it never blocks submission or hides an application from you. **Can I add more screening questions later?** Yes — `PUT` a revised definition; existing responses keep the `definition_version` they answered under, so past applications aren't retroactively rescored. **What if I don't want scoring at all?** Drop the `outcomes` block entirely — the rest of the form works fine as a plain application collector. --- Full schema, question types, and endpoints: [quizgen.dev/llms.txt](/llms.txt).