How to tailor your resume to a job description

Start with the posting, then your evidence.

Save the full job description before you edit. Separate required qualifications from preferred ones and from information about the company. A benefit such as a learning budget is not a skill you need to put on your resume.

Make a short working table with three columns: requirement, evidence in your experience, and where that evidence appears. If a requirement has no supporting evidence, leave that row empty. An empty row is useful information, not an instruction to invent a bullet.

• Required: building APIs. Evidence: the endpoints you implemented in a project.
• Preferred: PostgreSQL. Evidence: the database you actually used, if any.
• No evidence: managing a team. Leave it out if you have not managed one.

Change emphasis before changing sentences.

Move the most relevant supported bullet closer to the top of its role. Give a relevant project enough room to explain the work. Trim an unrelated detail if it crowds out evidence the reader needs.

Keep employers, dates, qualifications and job titles accurate. Tailoring is not permission to turn a contributor into a project lead or a class project into paid employment. Keep a main resume so each application starts from the same factual record.

Use the employer's language when it means the same thing.

If the posting asks for React and you built the interface in React, say React. If it asks for stakeholder communication and you presented findings to a client, describe the presentation and its audience. Naming a real activity is more useful than inserting a string of abstract keywords.

Do not list a tool merely because a similar tool appears in your experience. Someone who used MySQL has database experience, but that fact alone does not mean they used PostgreSQL. Explain the transferable work without replacing the tool name.

A rewrite should preserve every factual boundary.

Illustrative example, not a customer result: the original bullet is 'Built a React dashboard showing weekly support ticket counts from a CSV export.' The posting asks for data visualization and React interfaces. A supported rewrite is 'Built a React dashboard to visualize weekly support ticket counts from CSV exports.'

The rewrite makes the connection clear without adding a number, a promotion or a business outcome. 'Led an analytics team and cut support costs by 30%' is a different story. Nothing in the original supports it.

Read each new noun, number and claim of ownership as a question: where did this come from? If you cannot point to your own source material, remove it or replace it with a fact you can explain in an interview.

Review the actual file, not just the editing screen.

Open the exported resume. Check that the contact details are readable, the most relevant evidence has not been clipped, and dates still belong to the right roles. Copy a few lines into a plain-text editor to spot missing text or a surprising reading order.

Follow the employer's file instructions. A polished PDF is not useful if the application requests another format. Keep a copy named for the role so you can identify what you sent later.

A resume-tailoring prompt you can adapt.

Give the AI both the job description and your main resume. Remove personal details it does not need. Use the following prompt as a starting point, then verify the output against the originals.

Compare the job description with my resume. List the relevant requirements and the exact experience in my resume that supports each one. Flag requirements without evidence. Suggest reordered bullets and factual wording changes. Do not add skills, numbers, tools, job titles or responsibilities. For every edit, show the original, the revision and the source fact. Ask me about missing information instead of guessing.

Save the result as a separate version for this role. Check the actual exported file and every new claim. Litos brings the posting and tailored resume into one review workflow if you prefer that to copying material between a chat and a document editor.

Use a match score as a prompt to read.

A score can point to terms worth checking. It cannot certify that you meet every requirement, predict a recruiter's decision, or guarantee an interview. Ask what the score measures before you spend time chasing it.

Litos compares your resume with requirements from a posting and lets you review the tailored resume beside the job. The useful final check is still concrete: can a reader see the evidence, and is every claim yours?