karpathy-jobs-bls-visualizer

What This Skill Does

The karpathy-jobs-bls-visualizer is a powerful research utility designed to ingest, process, and visualize Bureau of Labor Statistics (BLS) Occupational Outlook Handbook data. It enables users to perform complex market analysis by layering various metrics—such as salary, education requirements, and growth outlooks—onto an interactive treemap. Beyond simple data visualization, the tool includes an LLM-powered scoring pipeline, allowing researchers to define custom criteria for evaluating AI exposure across 342 different occupations.

Installation

To deploy this skill, ensure you have the OpenClaw environment initialized. Run the following command in your terminal: clawhub install openclaw/skills/skills/adisinghstudent/karpathy-jobs-bls-visualizer

After installation, ensure you have your OPENROUTER_API_KEY set in your environment variables to enable the LLM scoring features. Follow the repository's provided instructions to sync dependencies and run the scraping pipeline.

Use Cases

• Labor Market Research: Quickly visualize which sectors are growing versus shrinking based on BLS projections.
• AI Impact Modeling: Define custom prompts to score how vulnerable or enhanced specific job categories are regarding humanoid robotics, generative AI, or automation.
• Career Planning: Use the treemap to filter occupations by salary and education level to find high-growth, high-value career paths.

Example Prompts

1. "Analyze the 342 occupations in the BLS dataset and score them based on susceptibility to remote automation using the karpathy-jobs-bls-visualizer pipeline."
2. "Update the treemap color layer to reflect high-salary, low-AI-exposure roles."
3. "Summarize the top 10 occupations with the highest growth potential for 2030 based on the current BLS processing output."

Tips & Limitations

• Data Integrity: The pipeline relies on raw HTML from the BLS; ensure your local html/ directory is populated using the provided scrape.py script before processing.
• Custom Scoring: The score.py script is modular. You can easily pivot from 'AI exposure' to 'Automation risk' or 'Physical task intensity' by simply updating the SYSTEM_PROMPT in the source code.
• Performance: Because the tool processes hundreds of pages, ensure your internet connection is stable during the initial scraping phase.