Career & Portfolio
How to Write Project Descriptions That Recruiters Actually Read
Recruiters give a project entry seconds. Lead with the problem, name the hardest part, and attach the proof — a four-line formula that turns a coursework list into a story worth interviewing about.
A recruiter reading your portfolio is not reading. They are scanning — a few seconds per project, looking for one reason to keep going or one excuse to stop. Most student project descriptions give them the excuse: a list of technologies, a sentence that begins "this is an app that...", and nothing about why the thing exists, what was hard, or whether it ever worked. The project might be genuinely good. The description makes it invisible.
The fix is not better adjectives. It is better information, in the right order. A strong project description answers four questions before the reader has to ask them — and it does it in about four lines.
The four-line formula
- The problem. One line: who has this problem, and what does it cost them? "Students lose track of who owes them money across paper notebooks and chat threads."
- What you built. One line: the thing itself, plainly. "A mobile app that records shared expenses and settles balances between friends."
- The hardest part. One line: the problem inside the problem, and how you solved it. "Keeping balances correct when two people edit the same record offline — solved with a double-entry ledger instead of editable totals."
- The evidence. One line: how anyone can verify it. A live link, a repository, a demo, a result. "Live demo and full source linked below; tested with automated checks on every core flow."
Four lines. Notice what is missing: the technology stack as the headline. Your stack belongs in the entry — recruiters and their filters do look for it — but as supporting detail, not as the opening argument. Nobody was ever hired because a description began with a framework name.
Start with the problem, not the stack
Compare two openings for the same project. First: "Built using React, Node.js and PostgreSQL. A web application for managing inventory." Second: "Small shops lose sales when stock runs out silently. I built an inventory system that warns before that happens." The first describes the author's tools. The second describes a person's problem and a reason the project exists. Only the second gives a recruiter something to remember — or something to ask you about in an interview, which is the entire point.
If you cannot write the problem line, that is diagnostic, not embarrassing: it usually means the project was built to learn a tool, which is a perfectly good reason — so say that honestly instead. "Built to learn how real-time systems work; the interesting part was..." is a stronger line than a pretend business case, because it leads straight to the hardest-part line, which is where your real skill shows.
The hardest part is the most valuable line
Every real project has a moment where it fought back: the bug that took two days, the design that collapsed and got rebuilt, the constraint you did not expect. Students hide these moments, assuming they make the project look weak. Recruiters and engineers read them the opposite way. Anyone can follow a tutorial to a working demo; the hardest-part line is the proof that you solved a problem nobody walked you through.
It is also your best interview material. Interviewers ask "tell me about a challenge in this project" because the answer reveals how you think. If that sentence is already in your description, you have rehearsed the story every time you have read your own portfolio — which is exactly how you talk about projects in an interview without rambling: the structure is already there.
Verbs beat adjectives
Description words like "robust", "seamless", "cutting-edge" and "user-friendly" carry no information, and experienced readers discount them on sight. Replace adjectives with verbs and specifics:
- Instead of "a robust authentication system" — write "email and Google sign-in with hashed passwords and rate-limited attempts."
- Instead of "a seamless user experience" — write "the full flow from signup to first result takes under two minutes."
- Instead of "an efficient algorithm" — write "replaced a full-table scan with an indexed lookup; the page stopped timing out."
- Instead of "worked in a team" — write "owned the API layer while two teammates built the mobile client."
Every replacement follows the same move: claim nothing, show something. If a sentence could be pasted onto any project in your class unchanged, delete it — it is saying nothing about yours.
Match every description to its proof
A description is a promise; the links under it are the evidence. Each project entry should point at the strongest proof you have: the live product if there is one, the repository if the code is the achievement, a short demo or write-up if neither is public. This is precisely what recruiters check in a portfolio — not whether the design is pretty, but whether claims survive one click. And the discipline scales: treating each repository as a case study instead of a code dump is the same habit, applied on GitHub.
Length is part of the discipline. If your description needs a scroll, the reader will not give it one — they will skim the first two lines and move on, which means your best material has to live at the top. Write the long version first if it helps you think, then cut it in half, then cut the adjectives that survived. What remains should be four lines a tired recruiter can absorb between two other tabs. The projects that get interviews are rarely the biggest ones in the portfolio; they are the ones whose value a stranger can understand in fifteen seconds.
A project description is not a summary of what you built. It is an argument for why a stranger should spend their next minute on you.
Rewrite it for every serious application
The four lines are a formula, not a script. For a frontend role, your hardest-part line might be the state-management rebuild; for a backend role, the same project's hardest part is the ledger design. Same project, different true story — chosen for the reader in front of you. Ten minutes of tailoring per application beats a hundred identical submissions, because the person reading can feel the difference between a description written for everyone and one written for them.
Pick your best project tonight and rewrite its description in four lines: the problem, the build, the hardest part, the evidence. It will feel uncomfortably short. That is the point — everything you cut was hiding the parts that were actually impressive.