The first version of a resume is the hardest one to write because the page is honest about what isn't there yet. The temptation is to inflate, to call a class project a "product", to call a summer internship a "leadership role", to round up. Most early-career advice leans into that temptation. This piece argues for the opposite move.
A strong first-role draft doesn't pretend the record is longer than it is. It selects, frames, and orders what's actually there until the case for fit is unmistakable. The reader on the other end has read enough of these to recognize inflation in two seconds. They haven't read enough to recognize composure in two seconds, but composure is the only thing that lands.
What counts as work
The first move is widening the definition of work without diluting it. Coursework counts when the assignment was real enough that someone outside the class would recognize the output. A senior project that shipped to a real client counts. A research paper that got cited counts. A class on machine learning where you trained five models on a dataset counts. A class where you read about machine learning and turned in three problem sets does not.
Side projects count when they have an audience or a measurable outcome. A side project with three users, a small revenue line, or a finished open-source release counts. A side project that exists only as a half-finished repo on GitHub does not, list it under interests if at all, not experience.
The early months of a job count, even if the job is current and short. Six months as a full-time engineer is six months of real work. Frame it as such. The trick is naming what you actually did inside those six months, not stretching it into a fiction of seniority.
How to order what's there
For a first real role, the conventional ordering (Education last, Experience first) is often wrong. A computer science senior with one strong capstone project and an internship has a stronger case made by leading with the capstone, especially if the internship was generic.
The honest ordering is whatever order leads with strength. If your strongest evidence is academic, lead with it. If it's a side project, lead with that. The resume is a case, not a chronology, and chronology is the wrong frame when the strongest material isn't the most recent. What counts as "leads with strength" for any given role comes from the JD itself, see reading the job description like an editor for the method.
Within each section, the same principle holds. Lead with the bullet that does the most work for fit, not the one that came first in time. A bullet about a measurable outcome (an experiment that moved a number, a project that shipped, a paper that got cited) earns the top of the entry over a bullet that names a responsibility.
A first-role resume is also the cleanest case for the one-page rule. The question of how long a resume should be has more nuance at senior levels, but for a first real role, the answer is one page, and the discipline of selection that comes with it is the entire point of this piece.
Education sits at the bottom; the senior capstone leads the experience section above the summer internship. The strongest case for fit is the most recent academic project, so it gets the top of the page.
The bullet as a sentence with one job
A first-role resume's bullets do one job each: they make a single claim about fit, with one piece of evidence behind it. Two-clause bullets fail. Bullets that try to summarize a whole project fail. The shape that works is I did X, which produced Y, where X is specific enough to picture and Y is concrete enough to verify.
Before
Worked on a team of four to develop a web application using React and Node.js for the senior capstone project.
After
Built the messaging layer of a four-person senior capstone in React, sustaining 40 concurrent connections in classroom testing.
Same length. The first names a context and a stack. The second names a contribution and what the code had to handle.
Where AI fits, and where it doesn't
A first draft of a resume is the wrong place to ask a model to fill in what isn't there. A model will give you words that sound like experience because words that sound like experience are everywhere in its training data. The bullets you get back will be confident, vague, and indistinguishable from every other bullet on every other resume the model has produced this week.
A second draft, against a specific role, is a different conversation. Once the underlying claims are yours and verifiable, refinement is mostly a question of cadence and emphasis, and that's the work tools like reecv are designed for.
Don't run the model on a blank page. Run it on a draft that's already yours.
Closing
The first resume is the hardest because the work to date is the smallest pile of material to draw from. That's not a writing problem; that's a career-stage problem, and pretending otherwise gets caught. The composed first draft (three or four entries, each one a real claim with real evidence, ordered for fit) wins more interviews than the inflated draft that tries to compete with senior records on volume.
Take the time the role is worth, name what you actually did, and let the page be honest about scale. The reader on the other end is paid to recognize fit, and fit doesn't require the largest pile.
Read next