How to Use Codex to Build a Job-Specific Resume

Most people paste their resume into ChatGPT, get something that sounds like everyone else’s resume, and call it done. Codex does something different. Instead of a single back-and-forth in a browser window, it works directly with your files — reading the job description, analyzing your experience, asking smart follow-up questions, and writing a tailored draft straight to your folder.

This guide walks you through the whole process, including the prompts that actually work.

Step 1: Choose Your Codex Client and Set Up a Workspace

Codex isn’t one thing. Depending on your plan and how you prefer to work, you can access it through the Codex app, the CLI (command-line interface), an IDE extension like VS Code, or the web. The workflow in this guide works across all of them — what matters is that you point Codex at a dedicated working directory so its useful context is concentrated there.

Create a Resumes parent folder if you don’t have one, then inside it create a subfolder for each job you apply to. Name each one something specific:

Resumes/
  └── acme-corp-marketing-director/
  └── spotify-brand-strategist/

Vague names like “resume-v2” or “job-copy” make it hard to pick back up mid-process and impossible to find six applications later. Be specific from the start.

Once you’ve opened Codex in that folder, any files it creates — your tailored resume, cover letter, change log — get written directly there. No copying from a chat window.

Privacy note before you go further: Your resume contains personal information — phone number, home address, sometimes more. Before you use any AI tool on a resume file, consider removing or replacing that information with placeholders. Use [phone] and [address] instead of the real values. You can add them back in the final version you send. This is especially worth doing if you’re experimenting with new tools or workflows for the first time.

Step 2: Prepare Your Files

This is where most people cut corners. Don’t.

Codex is only as good as what you feed it. Start with these files in your job folder before you run any prompts:

On the job description: Paste it straight from LinkedIn or a company careers page, or open a PDF and copy the text. Drop it into a new file in your folder and save it as job_description.md. Clean text is all Codex needs.

On your master resume: This is your “everything” document. Don’t filter it yet. Include roles you’ve de-emphasized, metrics you’ve forgotten to highlight, side projects — all of it. Let Codex figure out what’s relevant.

Step 3: Build Your Brag Sheet

Most resume guides skip this. It’s the most useful file in your folder.

Create accomplishments.md and fill it with raw notes about things you’ve done that you’d never put directly on a resume in their current form — rough, unpolished, specific. Things like:

- Led sponsor demos for clinical sample tracking software
- Helped translate vague client requirements into working configurations
- Worked with product, delivery, and technical teams across multiple accounts
- Supported pharma/CRO software sales conversations
- Built reusable demo narratives for complex workflows
- Covered for senior AE during Q3 — closed two deals solo

It doesn’t need to be organized. It doesn’t need to be grammatically clean. In fact, messy and specific is better than tidy and vague.

Your resume is the formal version of you. The brag sheet is the raw material Codex will mine to find things worth formalizing. Without it, Codex only has access to what you’ve already written — which means it’ll mostly just rearrange what you already have.

Step 4: Set Up Your Style Sample (Optional but Worth It)

If you have a resume you think is genuinely excellent — yours from a few years ago, a colleague’s, a template you found — don’t just hand it to Codex and say “match this.” Instead, run it through this prompt first to extract a proper spec:

Methodically reverse engineer this resume so I can recreate it with AI.
Extract everything — structure, section ordering, how accomplishments are
phrased, the balance of detail vs. brevity, formatting choices, and any
patterns in how information is prioritized. Be pixel-perfect accurate.

The words “methodically,” “reverse engineer,” and “pixel-perfect” aren’t filler — they push the AI to produce a detailed, usable rubric rather than a vague summary. Save that output as style_sample.md. Now when Codex writes your resume, it has a concrete standard to match, not just a vibe.

Step 5: Run the Gap Analysis First — Not the Resume

Here’s where this process separates from pasting into ChatGPT.

Do not ask Codex to write your resume yet.

Instead, make it read all three of your input files, compare them against the job description, and ask you follow-up questions before writing a single bullet point. The goal is to surface specific achievements from your brag sheet that your resume hasn’t captured yet — the kind of thing an AI can’t invent, and that you’ve probably undersold.

Paste this prompt into Codex:

Analyze resume_base.md, job_description.md, and accomplishments.md.

Your goal is to create the strongest truthful resume for this specific role.

First, do not write the resume.

Instead:
1. Identify the role's top priorities.
2. Compare those priorities against my current resume.
3. Search accomplishments.md for supporting evidence.
4. Flag anything that seems relevant but underdeveloped.
5. Ask me 3–5 follow-up questions that would make the resume stronger.

Do not invent metrics, tools, titles, or responsibilities.

Codex will come back with a structured analysis and questions. Answer honestly and specifically — the more detail you give here, the stronger the tailored draft will be.

Step 6: Generate the Tailored Draft

Once you’ve answered Codex’s questions, give it the green light:

Based on my answers and the gap analysis, write a tailored version of my
resume optimized for job_description.md.

Keep everything factually accurate — do not add metrics or achievements I
didn't confirm. Match the tone and formatting of style_sample.md if provided.

Save the output as resume_tailored.md.

The Save the output as instruction is important. Because Codex has file access, it writes the draft directly to your folder rather than spitting it into the terminal window. You get a file, not a wall of text to copy.

Step 7: Audit the Draft — Ask Codex to Show Its Work

This is the step most AI resume workflows skip entirely, and it’s what makes this approach auditable rather than just automated.

Before you accept the draft, run this prompt:

Before I finalize this resume, audit resume_tailored.md.

Create a change log that shows:
1. What you changed from resume_base.md
2. Why each change helps for this job description
3. Any claims that need my verification
4. Any phrasing that may sound inflated, generic, or too AI-written
5. Any missing evidence that would make the resume stronger

Codex will produce a structured breakdown of every significant change — what it rewrote, why, and where it’s uncertain. That gives you something to actually review, not just a draft to stare at and hope is correct.

This is the difference between “AI wrote my resume” and “AI helped me produce a job-specific version I can stand behind.”

Step 8: Review Before You Send

AI-generated resumes have a few common failure modes. Check for these before you send anything:

• Invented metrics — Codex should only use numbers you confirmed. If you see a figure you don’t recognize, it hallucinated it. Delete it.
• Inflated titles — Make sure your actual job titles are preserved exactly as they were. AI sometimes upgrades them.
• Tone mismatch — Does it sound like someone would actually say this, or does it sound like a LinkedIn auto-fill? If it’s the latter, read it out loud and rewrite the stiff parts.
• Keyword stuffing — Tailoring for a job description is good. Jamming every keyword into every bullet point is not. Read it as a human, not an algorithm.

Step 9 (Optional): Add a Cover Letter

If the role asks for one, Codex can draft it from the same context. Use this prompt after the resume is finalized:

Using resume_tailored.md and job_description.md, write a concise cover letter
for this role. 

Focus on why this specific role makes sense right now — not just a recap of
the resume. Address the hiring manager's likely pain points based on the JD.
Keep it under 300 words. Save it as cover_letter.md.

The “why now” framing matters. A cover letter that just restates the resume isn’t doing any work. The best ones explain the timing and the fit, which is exactly what Codex can infer from the job description context.

What This Won’t Do

A few honest caveats:

Codex works with what you give it. A thin brag sheet produces a thin resume. The gap analysis and follow-up questions help — but they can only surface what actually exists in your experience. Show up with real material.

This process also doesn’t format your resume visually. The output is a text file. You’ll paste it into your actual resume template — Word, Google Docs, a design tool — and apply your own formatting there.

And one more: the audit step in Step 7 catches a lot, but it doesn’t catch everything. Read the final draft out loud before you send it. If a sentence makes you wince, rewrite it yourself. Codex gets you most of the way there — the last 10% is yours.

Quick Reference

Next up in this series: how to tailor a resume using Claude — and when it’s actually the better tool for this job.