Show HN: I made "Who's Hiring?" searchable using GPT and Metabase

3 days ago 2

Fetches job postings from "Who is hiring?" threads on Hacker News, parses them with OpenAI, and stores them in Postgres for Metabase dashboards. Also exports to CSV.

A hn_jobs.csv file with all parsed jobs
Data in Postgres for Metabase visualization

Docker and Docker Compose installed and running
Python 3.9+
OpenAI API key

Clone:

git clone <your-repo-url> cd hn-jobs-to-metabase
Create .env:

POSTGRES_URL=postgresql://hnuser:hnpass@localhost:5432/hnjobs OPENAI_API_KEY=your_openai_api_key
Set thread ID: Edit MANUAL_THREAD_ID at the top of fetch_hn_jobs.py:

MANUAL_THREAD_ID = 43858554 # Set your thread ID
Run:

chmod +x run_local.sh ./run_local.sh
Metabase:
- Go to http://localhost:3000
- Connect to Postgres with these settings:
  - Host: db
  - Database: hnjobs
  - Username: hnuser
  - Password: hnpass
  - Schema: hn
- Explore the jobs table
Stop and Cleanup:

This will stop all containers and remove them, cleaning up your environment.

Search Hacker News for: Ask HN: Who is hiring? (Month Year)
The thread ID is at the end of the URL (e.g. item?id=43858554).
Paste it into MANUAL_THREAD_ID in fetch_hn_jobs.py.

CREATE SCHEMA IF NOT EXISTS hn; CREATE TABLE hn.jobs ( hn_comment_id bigint primary key, company text, role text, location text, salary text, contact text, description text, posted_at timestamp with time zone, created_at timestamp with time zone default now(), updated_at timestamp with time zone default now() );

MIT License

Read Entire Article