What this prompt does
This prompt runs your raw resume content through a structured rewrite pipeline built around the XYZ impact formula — "Accomplished X, measured by Y, by doing Z." That single constraint forces every bullet point to carry a result, not just a responsibility. Vague lines like "worked on backend APIs" become "Reduced API p99 latency by 38% by migrating synchronous DB calls to Redis-backed async queues."
Beyond bullet rewrites, it addresses ATS specifically — you supply the target role and key skills, and the output aligns your language to what parsing systems and hiring managers are actually scanning for. The technical projects section is the most differentiated piece: instead of listing a tech stack, it frames each project around a business problem, your individual contribution, and a measurable outcome. That shift from "what I built" to "what it changed" is what separates resumes that get callbacks from resumes that get archived.
When to use it
- You are applying for a senior or staff role and your current resume reads like a job description, not an achievement record
- You are switching from agency/freelance work to product companies and need to translate client work into product-relevant framing
- Your resume has passed zero ATS screens for a role you are clearly qualified for and you need a diagnostic plus rewrite
- You are applying to FAANG-adjacent companies where the bar for quantified impact in bullet points is explicit and well-documented
- You just shipped a major project and want to capture it while the metrics are fresh before it gets diluted into a generic line item
Example output
For a mid-level backend engineer targeting fintech startups:
**Senior Backend Engineer**
• Reduced payment processing failure rate from 3.2% to 0.4% by replacing synchronous
Stripe webhook handling with an idempotent queue-backed retry system (Laravel Horizon + Redis)
**Projects**
### Real-Time Fraud Detection Pipeline
Problem: Client's manual review team was reviewing 100% of flagged transactions —
8-hour average review time blocked legitimate users.
Contribution: Designed and implemented a scoring microservice in Go that consumes
Kafka events and applies rule-based + ML thresholds.
Stack: Go, Kafka, PostgreSQL, Docker, deployed on AWS ECS
Outcome: Automated 74% of decisions; avg review time dropped to 22 minutes
Pro tips
- Set
[experience_level]accurately — "mid-level" and "senior" generate different emphasis. Senior output leads with architecture decisions; mid-level leads with delivery and ownership - For
[key_skills], copy-paste 5-8 skill phrases verbatim from 3 real job postings you are targeting. ATS matching is literal, not semantic - Run the output back through the prompt a second time with a tighter
[target_role]after you see the first draft — iteration sharpens specificity - The red flags section is the most honest part of the output; read it before you edit the rewrites. It surfaces ordering problems, unexplained gaps, and overloaded tech lists that hurt more than help
- Pair this with a cover letter prompt that references the same
[company_type]variable — consistency between resume framing and cover letter framing is a signal most candidates miss
