Resume

📘 Learn Resume from zero

From zero: a resume is a one-page evidence summary whose only job is to convince a recruiter, in seconds, that you are worth a 30-minute phone screen. It is not your autobiography. Think of it as a movie trailer, not the full film: the trailer exists to make you buy a ticket, not to show every scene. The portfolio is the full film — opened only after the trailer worked.

Two first principles drive everything. (1) Attention is the scarce resource. A recruiter scans dozens of resumes and lingers a few seconds, mostly on the top third — so the strongest, most relevant evidence goes first. (2) Claims need proof and proportion. Each line should be an impact statement, not a duty. The canonical pattern, popularized by Google's hiring guidance (Laszlo Bock, Work Rules!), is the XYZ formula: Accomplished [X], as measured by [Y], by doing [Z] — i.e. result + metric + method.

Worked example. Weak: "Responsible for improving the checkout system." This is a duty with no scale and no outcome. Apply the formula: pick the accomplishment, the metric, and the method. Result: "Redesigned the checkout flow, lifting conversion ~12% by removing two form steps and adding Apple Pay." Now the reader sees scope, impact, and how — proof, not assertion. (Use a dollar figure only if you can defend the real number; an invented "$1.4M/yr" is worse than none.)

AI-assisted creation, done honestly. The XYZ pattern is exactly what makes a resume AI-assistable: paste your raw facts plus the job description and ask the model to rewrite each line into the formula and flag JD terms you are missing. But you supply every true number — the AI shapes wording, it does not source achievements. Watch for its failure modes: hallucinated metrics, interchangeable hype verbs ("spearheaded," "leveraged") with no Z, and keyword-stuffing that a human reads as noise.

Key insight: a resume is a relevance-and-evidence filter, not a record. Every line earns its space by being the most relevant provable impact for this specific role.

✨ Added by the guide to build intuition — not from the source course.

🎯 Guided practice

1. Easy — turn a duty into an impact bullet. Start: "Worked on the team's CI pipeline." Step 1, name the action: "Worked on" is vague, so upgrade to something concrete like Rebuilt or Automated (the X). Step 2, ask "what changed, measurably?" Suppose builds dropped from 20 min to 6 min (the Y). Step 3, ask "by doing what?" — parallelizing test stages and caching dependencies (the Z). Step 4, assemble with the XYZ formula: "Rebuilt the CI pipeline, cutting build time from 20 to 6 minutes (-70%) by parallelizing test stages and caching dependencies." Core lesson: every bullet must answer "what was the measurable outcome, and how?"
2. Medium — tailor one resume to a specific job. You have a generic resume and a posting that stresses "distributed systems, on-call ownership, mentoring." Step 1, extract the posting's top terms and rank them by how often and how prominently they appear — treat the job description as the spec your resume must match. Step 2, run a gap scan: for each top term, find a true bullet that proves it; if "mentoring" is required, absent, but real, add it. Step 3, reorder for the fast scan — move the distributed-systems and on-call bullets to the top of your most recent role so the highest-signal evidence lands in the top third. Step 4, mirror their literal vocabulary: if you wrote "ran the pager rotation" and they say "on-call ownership," adopt their phrase (truthfully) so both a human reader and the ATS parser register the match. Step 5, cut a low-relevance bullet to stay at one page — tailoring is as much deletion as addition. Core lesson: tailoring = match the JD spec with your strongest true evidence, ordered for a fast scan, never by inventing experience.

✨ Added by the guide — work these before the full problem set.