Why structure matters more than prose
AI extractors look for predictable section headings (Experience, Education, Skills, Projects). They look for date ranges in standard formats. They look for clear company names and job titles. The more you deviate from these conventions in service of "standing out," the more likely the extractor mis-categorizes you.
The good news: a well-structured resume gets parsed correctly, surfaces in more searches, and reads well to humans too. There’s no actual tradeoff.
Sections to include, in order
Contact info (name, email, phone, city, LinkedIn, portfolio). One line each.
Experience (most recent first). For each role: Company — Job Title — Dates — 3-5 bullets describing impact. Use numbers wherever you can.
Education. School, degree, graduation date, GPA only if 3.5+, relevant coursework if early-career.
Projects (especially for early-career). Each project: name, link, one-line description, the stack you used.
Skills. Group by category (languages, frameworks, tools). Don’t list "Microsoft Word."
Things to remove
Photos. Almost every US ATS rejects them silently. Some EU employers require them; check by country.
Two-column layouts. Many extractors read these in the wrong order.
Tables and text inside graphics. Extractors don’t read pixel content.
"Objective" statements. Replace with a one-line headline that names the role you’re looking for.
High-school information once you’re past sophomore year of college.