Most resumes have one or two great bullets, four or five passable bullets, and the rest is filler. The filler is the problem — it dilutes the strong bullets and makes the resume read average.
You don’t need to rewrite the whole resume. You need to find the weak bullets and rewrite the worst three.
The three weakness tests
Every bullet on your resume should pass three tests. Failing one is a yellow flag. Failing two is a rewrite. Failing all three is a delete.
Test 1: action verb leadership
The first word of the bullet should be a strong, role-appropriate action verb. Strong verbs: shipped, built, debugged, deployed, owned, drove, scaled, reduced, increased, designed, automated, eliminated.
Failing words: responsible for, tasked with, worked on, helped with, assisted in, participated in, involved with. These are filler verbs — they describe presence, not action.
If the role is execution-coded, lean execution verbs (ship, build, debug). If the role is strategy-coded, lean strategic verbs (drove, defined, aligned). The execution-vs-strategy verb conversion explains the swap.
Test 2: quantified outcome
The bullet should include a number, percentage, or concrete outcome. Not every bullet — about 60–70%.
Strong: “Reduced p99 latency from 480ms to 110ms (-77%) by replacing N+1 queries with a single batched join.” Weak: “Improved API performance significantly.”
The number does the work. “Significantly” does nothing — it’s a hedge.
When you can’t quantify, lead with a specific scope word: single-handedly, end-to-end, across three teams, on a 24-hour deadline. Specificity substitutes for numbers.
Test 3: role-specific signal
The bullet should map to a responsibility, skill, or signal from the JD you’re applying to. A generic bullet that fits any role fits no role well.
Run a check: pick a bullet, then re-read the JD. Does any phrase in the bullet correspond to a phrase in the JD? If not, the bullet is generic. Tailor it. Mirror the JD in your resume is the deeper read on phrase-mirroring.
How the diagnostic flags weak bullets
The diagnostic reads every bullet on your resume against the JD and runs all three tests in parallel. The output:
- Yellow bullets — failing one test, fixable in 30 seconds.
- Red bullets — failing two tests, queued for rewrite.
- Black bullets — failing all three, suggested cut or full rewrite.
For the bottom three weakest bullets, the diagnostic provides a before/after rewrite with the rationale: “Replaced ‘responsible for monitoring’ with ‘owned production observability;’ added ‘p99 latency from X to Y’ to quantify; mirrored ‘incident response’ from JD line 12.”
That’s the rewrite output you get from the diagnostic on the free tier.
Common weak-bullet patterns
The recurring patterns we see across thousands of resumes:
- Vague impact: “Improved customer satisfaction.” Replace with: “Reduced support ticket volume 38% by rewriting the onboarding flow.”
- Filler verb: “Responsible for code reviews.” Replace with: “Reviewed 200+ pull requests/quarter, blocking three critical regressions before merge.”
- Generic context: “Worked on the data pipeline.” Replace with: “Migrated the events pipeline from Kafka to Redpanda, cutting infra spend 40% with zero downtime.”
- Strategy-on-execution: “Aligned stakeholders on the new architecture.” On an IC role, replace with: “Designed and shipped the v2 architecture, then ran the migration across 12 services.”
The shape always changes. The work itself didn’t.
Run the bullet analyzer
Free, two cases, no card. Paste a JD, upload a resume, get every weak bullet flagged with a rewrite for the bottom three.
Related reading
- Rewrite resume bullets for a job description — the per-bullet rewrite framework.
- Execution verbs vs. strategic verbs — the verb-shape conversion.
- Mirror the job description in your resume — phrase-mirroring done right.
- Resume tuned for a specific job — the full per-application tailoring loop.