Filling a resume template and improving the wording with an LLM call is a genuinely simple combination — a form, a PDF template, and one well-crafted prompt.
Build it
Effort: an eveningRunning cost: $0/mo to start
Why CanICodeThis says Build it
A form collecting work history, education, and skills, rendered into a clean PDF template, is well-documented templating work
One LLM API call to tighten up bullet-point wording (given the job title and description) is a bounded, well-understood prompt pattern
Saving and editing multiple resume versions is ordinary CRUD
The parts that'll cause problems
Multiple polished template designs take real design iteration to look genuinely good, not just functional
Tailoring a resume automatically to a specific job posting (parsing the posting and re-weighting content) is a step beyond simple wording improvement
ATS (applicant tracking system) compatibility checking is a real, separate feature if you want it
How I'd build this
frontend
Next.js
database
Supabase
auth
Supabase Auth
hosting
Vercel / Cloudflare Pages
MVP scope
A form for work history, education, and skills
One clean PDF template
An LLM call to improve bullet-point wording
Save and edit multiple versions
Postpone to v2
Multiple template designs
Auto-tailoring to a specific job posting
ATS compatibility scoring
What it actually costs to build
AI agent — with a subscription (Claude Pro/Max, Cursor, etc.)
$0 marginal
AI agent — pay-per-use API, no subscription
$16–$27 one-time
Hosting, once it's running
$0/mo (free tier)
Domain name, if you want your own
~$12/yr
The build prompt
Build me a resume builder, genuinely achievable in a weekend:
1. A form for work history, education, and skills.
2. Render the form data into one clean PDF template using an existing PDF library (pdf-lib or a templating approach with Puppeteer).
3. A 'polish my bullet points' button sending the raw text plus a target job title to an LLM API, returning tightened wording you can accept or edit inline.
4. Save multiple resume versions per account.
That's a real, complete resume tool. Out of scope: multiple template designs, real additional layout work per template, and auto-tailoring to a specific job posting, which needs the posting's own text as additional LLM context — a real, separate feature, not a variation on polishing.