A developer named Santiago built an AI system called Career-Ops that evaluated 740+ job listings, generated 100+ tailored CVs, and helped him land a Head of Applied AI role. He then open-sourced it and it got 44,000+ GitHub stars in weeks. This guide shows you exactly how to set it up for free using Gemini AI (your Google account — no credit card needed).
Career-Ops is a free, open-source AI job search system built by Santiago Fernández, a former founder. After selling his business in 2025, he used AI to manage his entire job search — and the results were remarkable.
It turns your terminal into a complete job search command center. Instead of spending hours reading job descriptions, rewriting your CV, and filling application forms — Career-Ops does the analysis for you in seconds. You still make every final decision, but the AI filters the noise so you only spend time on jobs worth applying to.
| Task | Automated? | Notes |
|---|---|---|
| Score jobs A–F across 10 dimensions | ✅ Fully automatic | Evaluates fit against your CV |
| Generate tailored CV PDF per job | ✅ Fully automatic | ATS-optimized for each role |
| Write LinkedIn recruiter message | ✅ Fully automatic | Personalized per company |
| Give salary market insights | ✅ Fully automatic | Tells you negotiation range |
| Track all applications | ✅ Fully automatic | Saved in applications.md |
| Read job URLs automatically | ⚠️ Sometimes | LinkedIn blocks bots |
| Auto-fill and submit applications | ❌ Not supported | Most portals block automation |
Career-Ops is designed to filter jobs — not blindly apply to all of them. It recommends only applying to roles scoring 4.0/5 or above. Quality over quantity always wins.
Go to nodejs.org and click the LTS button (left side — "Recommended for most users"). Run the installer with all default settings.
Once installed, open Command Prompt (Windows key + R → type cmd → Enter) and verify:
node --versionGemini CLI is Google's free AI tool. It uses your personal Google account — no API key or credit card needed. In your Command Prompt, paste:
npm install -g @google/gemini-cliThen launch it to log in with your Google account:
gemini authA browser window will open — sign in with your personal Google account and click Allow. Come back to the terminal once done.
Run these commands one by one in your Command Prompt:
# Go to Documents folder
cd Documents
# Download Career-Ops
git clone https://github.com/santifer/career-ops.git
# Go into the folder
cd career-ops
# Install dependencies
npm install
# Install PDF generator (takes 2-3 mins)
npx playwright install chromiumRun these two commands to create your config files from templates:
copy config\profile.example.yml config\profile.yml
copy templates\portals.example.yml portals.ymlOpen File Explorer → Documents → career-ops → create a new file called cv.md. Open it in Notepad and paste your CV in this format:
# Your Full Name
**Email:** your@email.com
**Phone:** +91-XXXXXXXXXX
**Location:** City, India
**LinkedIn:** linkedin.com/in/yourprofile
## Summary
2-3 lines about yourself and target roles.
## Experience
### Job Title — Company Name (Year - Year)
- What you did
- Key achievement
- Responsibility
## Skills
- Skill 1, Skill 2, Skill 3
## Education
**MBA** — College Name (Year)npm run doctorMake sure you're inside the career-ops folder, then launch Gemini:
cd Documents\career-ops
geminiWhen it asks about trusting the folder — select option 1 "Trust folder" and press Enter. You'll see 1 GEMINI.md file · 1 skill at the bottom — this confirms Career-Ops is loaded.
Find a job on LinkedIn, Naukri, or any company portal. Copy the job URL or the full job description text. Then type in Gemini:
/career-ops-evaluate https://www.linkedin.com/jobs/view/XXXXXXXXXIf LinkedIn blocks the URL (very common), paste the job description text directly instead:
/career-ops-evaluate
[paste full job title, company, and description here]✅ Match score (e.g. 4.6/5 — Strong Match)
✅ Tailored PDF CV saved to output/ folder
✅ LinkedIn recruiter message ready to send
✅ Salary market range for the role
✅ Application tracked in data/applications.md
Once Gemini is running inside the career-ops folder, these are all the commands you can use:
| Command | What it does |
|---|---|
| /career-ops-evaluate [URL or JD] | Full evaluation — score + CV + tracker + recruiter message |
| /career-ops-scan | Scan job portals for new matching offers automatically |
| /career-ops-batch | Evaluate 10+ jobs at the same time in parallel |
| /career-ops-tracker | View all your applications and their status |
| /career-ops-pdf | Generate ATS-optimized CV PDF for current job |
| /career-ops-deep | Deep research on the company before applying |
| Draft a LinkedIn message to the recruiter | Write a personalized outreach message |
| Prepare me for a potential interview at [company] | Generate interview prep tips |
LinkedIn actively prevents automated access. Instead of using LinkedIn URLs, go to the job page, copy the full job description text, and paste it directly into Career-Ops. Naukri.com and company career pages work much better.
Career-Ops evaluates jobs on 10 dimensions. The system itself recommends not applying to anything below 4.0. Your time is valuable — spend it only on strong matches.
Career-Ops learns who you are through your CV and the onboarding questions. Give it detailed answers about your superpower, deal-breakers, and best achievements. The more context you give, the better it evaluates jobs for you.
Every evaluation includes market salary data for that role in your city. Many Indian job seekers undervalue themselves — use this data to negotiate confidently. Career-Ops often shows ranges ₹5-10L above what candidates initially expected.
Career-Ops is a local tool. Your CV, personal details, and job search data stay on your machine. It only sends text to Google's Gemini API — nothing is stored on any Career-Ops server.
Using this exact setup, we evaluated a Guidewire PMO Senior Consultant role and got:
Score: 4.6/5 — Strong Match · Salary insight: ₹22L–₹32L+ (above the ₹15-20L target) · Tailored PDF CV generated automatically · LinkedIn recruiter message written and ready to send — all in under 5 minutes.