I’ve been counting public tech job ads every week and publishing the counts as CSVs. It’s three weeks in now, so I’m sharing it in case it’s useful to someone.<\/p>\n
Each week gets a folder with three files:<\/p>\n
– job_types.csv: jobs counted, jobs long enough to read, and employers, for 5 job types (AI, machine learning, DevOps, software engineering, entry level software)<\/p>\n
– skills.csv: which skills the ads name, how many list each one as required, and the share<\/p>\n
– experience.csv: years of experience required, for jobs with an early career title and for all software jobs<\/p>\n
How it’s counted: ads from about 255 employer boards (Greenhouse, Ashby, Lever, Workable) plus a few job sites. The same title at the same employer counts once. Skills are matched with fixed rules against a list of about 250, no language model.<\/p>\n
Caveats<\/strong>: the boards lean towards US employers, AI companies and fast growing firms, so it’s not the whole market. It counts ads, not hires. Three weeks isn’t a trend yet. It updates every Monday.<\/p>\n Latest week<\/strong>: https:\/\/github.com\/DerrickWawerumuturi\/glassbox-job-counts\/tree\/main\/2026-10-19<\/a><\/p>\n I also wrote up one thing it shows (629 of the 4,536 software jobs that week were entry level, and 274 of those were internships): https:\/\/seeglassbox.com\/market\/entry-level-software\/is-entry-level-dead?utm_source=reddit&utm_campaign=datasets<\/a><\/p>\n Disclosure<\/strong>: I made this, it’s my project (Glassbox). The data is CC BY 4.0 and the story needs no sign up. I’d like to hear what you’d want counted next.<\/p>\n<\/div>\n submitted by \/u\/dw_mt <\/a>
[link]<\/a><\/span> [comments]<\/a><\/span><\/p>
