Key takeaways
- India has more geo-resolved GitHub developers tagged ml-ai than the United States — 4,810 versus 3,536 — a reversal of the overall GitHub population, where the US leads India 43,996 to 32,915.
- 28,603 of the 972,576 profiles in Crawlora's GitHub Users census (2.9%) carry the ml-ai interest-domain tag, inferred from repo topics and bio text.
- ML/AI-tagged developers are dramatically more reachable than the general GitHub population: 56.5% expose a public contact channel versus 21.2% overall, and 21.1% mark themselves open to hire versus 7.5% overall.
- Academic institutions crack the top employer/affiliation mentions for this group in a way they don't for GitHub overall — Northeastern University, Stanford, CMU, Tsinghua and Peking University all appear in the top ranks, alongside Microsoft, NVIDIA, Red Hat and Google.
- Follower counts skew higher too: 90.5% of ML/AI-tagged developers sit in the lowest (nano, <100) follower tier, versus 95.3% for GitHub overall — the whole distribution shifts up a notch, though the single most-followed ml-ai account (38,746 followers) still falls short of GitHub's overall top accounts.
Our census of 972,576 GitHub developer profiles found the US narrowly ahead of India in raw developer count. Filter that same dataset down to the 28,603 profiles tagged ml-ai — Crawlora's inferred interest-domain label for AI/ML-focused developers — and the ranking flips: India has more geo-resolved AI/ML developers than the United States.
Method at a glance
- Same census, one filter: the same 972,576-profile GitHub Users dataset behind our earlier study, filtered to
domain=ml-ai— an interest-domain tag inferred from a developer's repo topics and bio text, not a self-declared label.- 28,603 profiles carry the tag (2.9% of the full census). Geography figures below are over the subset of those that geo-resolve to a country/city, same caveat as the parent study: most developers don't publish a location at all.
- Total vs. queryable: 972,576 total indexed profiles; a default search returns 970,617 — the 1,959 accounts flagged as suspected automation (mass-repo commit farms) are excluded by default. Baseline percentages compared against "GitHub overall" below use that 970,617 queryable base, the same denominator as the parent study.
- Reachability and hireability are computed the same way as the parent study — a public contact channel (email, Twitter/X, blog, or social account) and GitHub's own "available for hire" flag, respectively.
- Snapshot dated July 2026, pulled live from Crawlora's GitHub Users API.
The geography flips: India leads AI/ML, not the US
Across the whole GitHub census, the US leads India in raw developer count. Filtered to ml-ai, India pulls ahead:
By city, Bengaluru tops the list at 518 — consistent with it also leading the overall GitHub census — but the rest of the top cities show the same India-heavy pattern the country totals do: Hyderabad (289), New Delhi (254), Pune (235) and Chennai (218) all place in the global top 10, alongside San Francisco (337), New York (305) and London (304).
Show the top 25 cities by ml-ai-tagged developer count
Top cities by geo-resolved GitHub developers tagged ml-ai
| City | Developers |
|---|---|
| Bengaluru | 518 |
| San Francisco | 337 |
| New York | 305 |
| London | 304 |
| Hyderabad | 289 |
| New Delhi | 254 |
| Pune | 235 |
| Chennai | 218 |
| Paris | 192 |
| Seoul | 178 |
| Dhaka | 164 |
| Mumbai | 161 |
| Toronto | 155 |
| Berlin | 148 |
| Beijing | 143 |
| Kolkata | 138 |
| Seattle | 128 |
| Lahore | 127 |
| Karachi | 104 |
| Boston | 101 |
| Noida | 99 |
| Coimbatore | 97 |
| Los Angeles | 95 |
| Nairobi | 91 |
| Islamabad | 87 |
Top 25 cities by geo-resolved ml-ai-tagged GitHub developers, July 2026. Source: Crawlora GitHub Users dataset, domain=ml-ai filter.
Academia shows up here in a way it doesn't for GitHub overall
Our earlier study found Red Hat and Microsoft topping the employer mentions for GitHub overall, with zero universities in the top 10. Filtered to ml-ai, universities crack the list outright:
Microsoft, NVIDIA and Red Hat lead — the same pattern of open-source-forward companies from our earlier study — but Northeastern University (16 mentions) alone outranks SAP, Deloitte, ByteDance and Huawei among named affiliations, and Stanford, CMU, Tsinghua, Peking and Zhejiang all appear further down the list. AI/ML research still runs substantially through universities, and their researchers' GitHub bios reflect it in a way the general developer population's don't.
This population is dramatically easier to reach and recruit from
The starkest difference between ml-ai-tagged developers and the general GitHub population isn't geography — it's reachability:
56.5% of ml-ai-tagged developers expose some public contact channel, versus 21.2% of the general population — and 21.1% mark themselves open to hire, versus 7.5% overall. Both figures run roughly 2.7-2.8x higher than the GitHub-wide baseline. Activity levels don't explain the gap (93.9% of ml-ai-tagged developers pushed, opened a PR, or reviewed one in the last 90 days, essentially matching the overall population's 93.6%) — this looks like a hiring-market effect: developers building visibly in a hot field have more reason to make themselves findable.
Follower distribution shifts too, though more modestly. 90.5% of ml-ai-tagged developers sit in the nano tier (under 100 followers), versus 95.3% of GitHub overall; the micro tier (100-999) holds 8.3% versus 4.2% overall. The single most-followed developer carrying the ml-ai tag in this dataset — Sebastian Raschka (@rasbt, author of several ML textbooks) at 38,746 followers — doesn't reach GitHub's overall macro-tier leaders, a reminder that broad platform fame and deep specialist reputation are correlated but distinct.
What this means if you're hiring, sponsoring, or building AI tooling
For technical recruiting in AI/ML specifically, India isn't a secondary sourcing market — by this measure it's the largest single geography, ahead of the US. For OSS sponsorship and DevRel aimed at the AI/ML community, the 2.7x reachability gap means outreach lists built from this population will convert on contactability far better than a generic GitHub-wide list would. And for academic partnership or research-collaboration tooling, university affiliations are a real, measurable signal in this specific population in a way they simply aren't for GitHub developers generally.
For a look at how a different specialist population compares, see GitHub's security researchers — the US leads there instead of India, and the follower-tier skew is even steeper than what we found here.
Sources
- Crawlora — GitHub Users dataset
- Crawlora — Where the World's Developers Actually Live (parent study)
- Crawlora — GitHub API docs
Frequently asked questions
Does India or the US have more AI/ML developers on GitHub?
India, by this measure. Filtering Crawlora's 972,576-profile GitHub Users census to the 28,603 profiles tagged ml-ai (an interest-domain tag inferred from repo topics and bio text), India has 4,810 geo-resolved developers versus 3,536 for the United States - a reversal of the overall GitHub census, where the US leads India 43,996 to 32,915.
How many GitHub developers work in AI/ML?
28,603 of the 972,576 profiles in Crawlora's GitHub Users dataset (2.9%) carry the ml-ai interest-domain tag. This is inferred from repo topics and bio text, not a self-declared label, so it should be read as a lower-bound signal rather than an exact headcount.
Are AI/ML developers easier to reach and recruit than other GitHub developers?
Yes, substantially. 56.5% of ml-ai-tagged developers expose a public contact channel, versus 21.2% of GitHub overall - and 21.1% mark themselves open to hire, versus 7.5% overall. Both figures run roughly 2.7-2.8x higher than the GitHub-wide baseline, while activity levels (93.9% active in the last 90 days) barely differ from the overall population's 93.6%.
Do universities show up in AI/ML developers' GitHub profiles?
Yes, in a way they don't for GitHub overall. Northeastern University, Stanford, Carnegie Mellon, Tsinghua and Peking University all appear among the top named affiliations for ml-ai-tagged developers, alongside companies like Microsoft, NVIDIA, Red Hat and Google. The overall GitHub census's top 10 employer mentions include zero universities.
Originally published on crawlora.net. Crawlora is a structured web-data, search, and anti-bot API — dozens of platforms as normalized JSON, plus a hosted MCP server, with a free tier (no card).
Top comments (0)