GitHub developer dataset: reachability and availability by city
This is a free, aggregated dataset of 82,483 active GitHub developers across the United States, France and the United Kingdom, broken down by metropolitan area, technology and engineering role. For every group it reports how many developers it contains, what share publish an email address on their profile, and what share have flagged themselves as available for hire. It contains no individual records and is published under CC BY 4.0.
What the numbers say
- Public email addresses vary widely between cities. In Raleigh, 46% publish an email address on their profile; in Nantes, 29%.
- Open-to-work flags vary more than twofold between cities. In Miami, 29% flag themselves as open to work; in Bristol, 13%.
By country
| Country | Developers | Public email | Open to work |
|---|---|---|---|
| United States | 57,728 | 39% | 22% |
| United Kingdom | 13,092 | 34% | 21% |
| France | 11,663 | 31% | 18% |
By city
Cities with at least 50 indexed developers. The full file goes further, down to city and technology combinations.
| City | Country | Developers | Public email | Open to work |
|---|---|---|---|---|
| San Francisco | United States | 9,252 | 40% | 21% |
| London | United Kingdom | 6,666 | 35% | 21% |
| New York | United States | 5,424 | 36% | 21% |
| Seattle | United States | 4,308 | 39% | 18% |
| Paris | France | 4,307 | 32% | 18% |
| Boston | United States | 2,726 | 40% | 19% |
| Los Angeles | United States | 2,178 | 39% | 21% |
| Chicago | United States | 1,688 | 38% | 21% |
| Austin | United States | 1,643 | 40% | 22% |
| Denver | United States | 1,598 | 39% | 21% |
| Atlanta | United States | 1,277 | 36% | 22% |
| Portland | United States | 1,170 | 44% | 22% |
| Washington | United States | 991 | 38% | 23% |
| Dallas | United States | 858 | 40% | 26% |
| San Diego | United States | 826 | 39% | 19% |
| Philadelphia | United States | 778 | 41% | 23% |
| Lyon | France | 748 | 33% | 16% |
| Manchester | United Kingdom | 596 | 33% | 18% |
| Raleigh | United States | 592 | 46% | 20% |
| Phoenix | United States | 578 | 42% | 27% |
| Minneapolis | United States | 568 | 41% | 20% |
| Pittsburgh | United States | 568 | 36% | 16% |
| Toulouse | France | 516 | 31% | 21% |
| Houston | United States | 512 | 39% | 28% |
| Detroit | United States | 506 | 41% | 21% |
Method, and what this does not measure
Figures come from public GitHub profiles and repositories, enriched and deduplicated. A developer counts once per group. Locations are free text on GitHub, so they are normalised to a metropolitan area before counting, and suburbs fold into their hub.
Two limits worth stating. This measures the developers we have indexed, not every developer in a city: a place with fewer rows may simply be less covered. And "open to work" is GitHub's own self-declared flag, which people set and then forget, so read it as a floor rather than a census.
Groups smaller than 50 developers are left out entirely. Below that, a percentage is rounding noise, not a measurement.
Reuse
Published under CC BY 4.0. Use it anywhere, including commercially, as long as you credit StarHunt and link back.