Career-Ops - Open-Source AI Job Search System, Automatically Generates Customized Job Application Materials

What is Career-Ops

Career-Ops is an open-source AI-driven job search system built on Claude Code. It is not a traditional "mass application tool" but an intelligent filter that helps job seekers screen out opportunities truly worth investing effort in from a massive number of job postings through structured evaluation, and automatically generates customized job application materials. The author used this system to evaluate over 740 job positions, generate over 100 ATS-optimized resumes, and ultimately successfully landed a Head of Applied AI position.

Main Features of Career-Ops

• Intelligent Job Evaluation: After pasting a job URL or description, it automatically performs a six-dimension analysis (role summary, resume match, level strategy, compensation research, personalization, interview preparation) and outputs a structured report.
• ATS Resume Generation: Dynamically adjusts resume keywords and experience presentation based on the job description, uses Playwright to render PDFs, and includes built-in Space Grotesk and DM Sans fonts to ensure passing applicant tracking systems screening.
• Portal Scanner: Pre-configured with 45+ companies (including Anthropic, OpenAI, ElevenLabs, etc.), automatically scrapes job postings from platforms such as Greenhouse, Ashby, Lever, and Wellfound via Playwright, and supports custom queries.
• Batch Parallel Processing: Uses Claude sub-agents to evaluate more than 10 job positions in parallel, greatly improving screening efficiency.
• Interview Story Library: Automatically accumulates behavioral interview material in STAR+Reflection format during the evaluation process, gradually building 5-10 reusable core story templates.
• Negotiation and Outreach: Provides salary negotiation frameworks, geographic discount rebuttal strategies, competitive offer leverage scripts, and LinkedIn outreach message templates.
• Terminal Dashboard: A TUI interface built on Go + Bubble Tea, supporting 6 filter tags and 4 sorting methods, enabling full-process management within the terminal.

How to Use Career-Ops

• Clone the repository and install dependencies：git clone https://github.com/santifer/career-ops.git + npm install + npx playwright install chromium
• Run environment check：npm run doctor
• Configure personal information：cp config/profile.example.yml config/profile.yml, fill in personal background, target salary, tech stack, etc.
• Create a resume file: Create in the project root directory cv.md, write Markdown-format resume content
• Start using: Launch Claude Code for personalized tuning, then paste a job URL or run /career-ops to trigger the complete pipeline

Career-Ops Official Website Address

• Github Repository：https://github.com/santifer/career-ops

Key Information and Usage Requirements of Career-Ops

• Strong Dependence on Claude Code: Requires a paid Anthropic subscription to use the core AI features.
• Environment Requirements: Node.js + npm + Playwright + Go (for Dashboard).
• Data Storage: Purely local storage, no cloud upload, privacy controllable.
• Main Market: Currently mainly targeting the English/European and American tech market; adaptation for Chinese and non-technical positions requires self-adjustment.
• Compliance Reminder: Automated scraping must comply with the terms of service of each recruitment platform; AI-generated content must be manually reviewed to avoid integrity risks.
• Learning Curve: Initial evaluation accuracy is limited, and continuous context feeding is required to improve matching quality.

Core Advantages of Career-Ops

• Anti-Spray-and-Pray Strategy: The system explicitly opposes the "casting a wide net" approach to job hunting and uses an A-F scoring system to score positions across 10 weighted dimensions (salary match, tech stack fit, growth potential, etc.), with a maximum score of 5.0. It strongly recommends not applying to positions scoring below 4.0.
• Human-in-the-Loop: AI handles analysis, recommendations, and material preparation, but the system will never automatically submit applications. The final decision-making power always remains with the user.
• Self-Evolution Capability: Supports using natural language to directly ask Claude to modify configuration files, such as adjusting scoring weights, career archetypes, target company lists, etc., enabling system self-customization.
• Plain-Text Data Layer: All data is stored as Markdown tables + YAML configuration + TSV batch files, supports Git version management, and is fully transparent and controllable.
• MIT Open-Source License: Completely free, community-driven, with auditable code that can be forked and further developed.

Comparison of Career-Ops with Similar Competitors

Application Scenarios of Career-Ops

• Large-Scale Initial Job Screening: When facing dozens or hundreds of new positions each week, quickly filter out roles with low match.
• Customized Resume Submission: For high-match positions, generate an ATS-friendly customized resume with one click.
• Systematic Interview Preparation: Start building a library of behavioral interview stories in STAR format during the application stage.
• Salary negotiation support: After receiving an offer, use the system's built-in negotiation framework and scripts to negotiate salary.
• Long-term career planning: Through reviewing scoring data, analyze your market positioning and trends in competitiveness.