A working student built an automated pipeline using dlt and Claude to analyze 987 HackerNews 'Who is Hiring?' job posts across 3 months (Apr–Jun 2026). The pipeline extracts tools, roles, locations, and remote status from job listings using Claude Haiku, runs incrementally (first monthly run ~$0.40, reruns free), and auto-updates a live dashboard. Key findings: Python and AWS dominate, Claude is now mentioned in 5.3% of job posts (up from 3.6% in April), while Copilot mentions are declining. The whole project was built in one AI conversation using dlt AI Workbench and deployed with a single command.
1.1K Impressions