Developer Talent Scarcity Index
Which tech stacks are hardest to hire for in the US? We compared 30 days of US job postings (489,940 technology mentions) with the number of developers who actually use each technology.
Key findings
114,352 US postings from 21,468 companies against about 1.24M developers. Much of the pull comes from data and AI roles.
AWS, Kubernetes, Terraform are all severe, and their developers earn about 15% above the US median ($155,000).
Median framework: 5.2 postings per 1K developers, against 22 for languages and 41 for cloud & infrastructure.
24% of developers who don't use it want to, and 70% of users want to keep using it.
The index
Click a column to sort. Each technology links to its stack page.
| # | Technology | Score | Roles / 1K devs | Postings (30d) | US devs (est.) | Pay vs median | Retention | Demand shift |
|---|---|---|---|---|---|---|---|---|
| 1 | Python | 100 Severe | 92.6 | 114,352 | 1,235,300 | +3% | 52% | +1% |
| 2 | Java | 91 Severe | 63.9 | 36,662 | 573,500 | +3% | 34% | -10% |
| 3 | AWS | 90 Severe | 61.1 | 70,717 | 1,158,300 | +10% | 50% | -6% |
| 4 | C++ | 90 Severe | 61.0 | 29,356 | 480,900 | -3% | 44% | +8% |
| 5 | Scala | 89 Severe | 56.2 | 3,614 | 64,300 | +30% | 35% | -8% |
| 6 | Kubernetes | 84 Severe | 46.0 | 29,414 | 639,600 | +16% | 52% | +2% |
| 7 | .NET * | 84 Severe | 46.9 | 33,837 | 721,500 | -10% | 60% | +67% |
| 8 | Terraform | 81 Severe | 41.0 | 18,714 | 456,700 | +19% | 49% | +4% |
| 9 | Go | 75 Severe | 31.4 | 12,695 | 404,800 | +16% | 52% | +7% |
| 10 | PHP | 71 High | 26.5 | 7,675 | 289,100 | -16% | 34% | -23% |
| 11 | React | 63 High | 18.5 | 19,024 | 1,030,300 | +3% | 48% | -11% |
| 12 | JavaScript | 62 High | 17.8 | 25,959 | 1,457,700 | -3% | 42% | -10% |
| 13 | TypeScript | 62 High | 17.9 | 17,904 | 998,700 | +3% | 55% | +10% |
| 14 | Ruby | 60 High | 16.2 | 3,362 | 207,100 | +13% | 44% | -16% |
| 15 | Swift | 60 High | 16.5 | 2,416 | 146,700 | +8% | 52% | -34% |
| 16 | Rust | 59 High | 15.3 | 5,378 | 351,900 | +13% | 70% | +32% |
| 17 | Docker | 58 High | 15.1 | 21,922 | 1,451,700 | +6% | 58% | -5% |
| 18 | Node.js | 54 High | 12.8 | 13,612 | 1,063,800 | +1% | 45% | -3% |
| 19 | Kotlin | 53 High | 12.2 | 2,475 | 202,400 | +11% | 47% | -18% |
| 20 | PostgreSQL | 52 High | 11.5 | 12,941 | 1,130,000 | +8% | 65% | +1% |
| 21 | Flutter * | 47 Moderate | 9.3 | 711 | 76,100 | +10% | 47% | -35% |
| 22 | Django | 41 Moderate | 7.1 | 1,600 | 224,700 | +3% | 42% | -16% |
| 23 | Next.js | 36 Moderate | 5.7 | 2,284 | 402,800 | +3% | 36% | -15% |
| 24 | Ruby on Rails | 32 Moderate | 4.8 | 936 | 196,400 | +13% | 50% | -30% |
| 25 | Vue | 24 Low | 3.4 | 1,097 | 326,300 | -5% | 48% | -22% |
| 26 | Elixir | 18 Low | 2.6 | 182 | 70,600 | +16% | 67% | -5% |
| 27 | Laravel | 15 Low | 2.3 | 263 | 114,700 | -13% | 51% | -28% |
| 28 | Angular * | 5 Low | 1.5 | 625 | 417,500 | -3% | 37% | – |
| 29 | Svelte | 0 Low | 1.2 | 213 | 170,800 | -6% | 64% | +3% |
Score 0–100, higher = harder to hire. Pay vs median: what developers using it earn relative to all US developers. Retention: share of current users who want to keep using it. Demand shift: change in its share of all postings since April 2026. * .NET: Supply = C#, F#, VB.NET, ASP.NET, ASP.NET Core or Blazor. Flutter: Supply measured via Dart usage (the survey has no Flutter option). Angular: Supply and demand both include AngularJS and Angular 2+.
Scarcity at a glance
Open US postings per 1,000 developers who use each technology.
What it means for your hiring plan
Severe Expect long searches and bidding wars
Demand far outstrips the qualified pool. Budget for a premium or a long time-to-fill. Contractors and specialist agencies are often the fastest way to get senior people on the work while a permanent search runs.
Python, Java, AWS, C++, Scala, Kubernetes, .NET, Terraform, Go
High Hireable, but plan for it
Plenty of developers know these, but competition for senior people is real. Direct hiring works if you move quickly; bring in contract help for deadlines you can't move.
PHP, React, JavaScript, TypeScript, Ruby, Swift, Rust, Docker, Node.js, Kotlin, PostgreSQL
Moderate Healthy supply
The talent pool comfortably covers current demand. Standard hiring processes should work, and freelance and agency rates stay competitive.
Struggling to hire for a scarce stack?
We take on contract development work, so you can start now while the permanent search runs. Tell us the stack and the problem and we'll reply by email.
- MVPs and new products
- Developers for your existing team
- Rescuing stalled projects
- AI features
Where to find adjacent talent
Less-scarce, skill-adjacent groups with developers who already want to move into each scarce skill: good candidates for cross-training or a stretch hire.
| Scarce skill | Recruit from |
|---|---|
| Python Severe | Node.js developers (15% want to move into Python); Java developers (13% want to move into Python); Ruby developers (9% want to move into Python); Go developers (6% want to move into Python) |
| AWS Severe | Kubernetes developers (11% want to move into AWS); Docker developers (9% want to move into AWS); Terraform developers (9% want to move into AWS) |
| C++ Severe | Rust developers (6% want to move into C++); Go developers (5% want to move into C++) |
| Kubernetes Severe | Terraform developers (14% want to move into Kubernetes); Docker developers (14% want to move into Kubernetes) |
| .NET Severe | TypeScript developers (6% want to move into .NET) |
| Terraform Severe | Docker developers (8% want to move into Terraform) |
| Go Severe | Node.js developers (24% want to move into Go); Rust developers (17% want to move into Go) |
| React High | Angular developers (13% want to move into React); Vue developers (10% want to move into React) |
| TypeScript High | Node.js developers (20% want to move into TypeScript) |
| Swift High | Kotlin developers (5% want to move into Swift) |
| Docker High | Node.js developers (8% want to move into Docker) |
Developers who use the source technology but not the target, and say they want to work with the target next year. Pools under 150 respondents and interest under 5% omitted.
Reading between the lines
Among harder-to-hire technologies, Java (34%), PHP (34%), Scala (35%) have the lowest retention: fewer than half of current users want to keep working with them. For a long-lived product that means an ageing, shrinking talent pool and rising maintenance costs. If hiring speed matters for something new, favour stacks with a large pool, moderate scarcity and strong retention, such as PostgreSQL (65% retention) and TypeScript.
Frequently asked questions
Which tech stack is hardest to hire for in the US?
Python, with about 93 open job postings for every 1,000 US developers who use it, followed by Java (64) and AWS (61). Figures cover the 30 days to September 30, 2026.
Which tech stacks are easiest to hire for?
Svelte, Angular and Laravel have the most available developers relative to open roles, at under 2 postings per 1,000 developers.
Should I hire full-time or use contractors for a scarce stack?
For Severe-tier stacks a permanent search can take months. Many teams bring in contractors or a specialist agency to start immediately while hiring runs in parallel. For Moderate and Low tiers, direct hiring is usually fast and cost-effective.
How is the scarcity score calculated?
30-day US job postings mentioning each technology (TheirStack), divided by the estimated number of US professional developers who use it (Stack Overflow Developer Survey 2025 usage share, scaled to the BLS developer workforce), placed on a log scale and normalised to 0-100.
Methodology
- Demand: TheirStack US job postings, 30-day window, collected September 30, 2026. A posting counts for every technology it mentions.
- Supply: Stack Overflow Developer Survey 2025, US professional developers (5,531 respondents): the share who used each technology in the past year, scaled to a US developer workforce of 2,037,050 (BLS OEWS May 2025: software developers, QA/testers, web developers, programmers).
- Score: postings per 1,000 estimated developers, log-scaled and normalised to 0–100. Tiers: Severe ≥ 75, High ≥ 50, Moderate ≥ 25, Low < 25.
- Demand shift: share of all tracked postings now versus April 28, 2026. We compare shares because the job-data provider keeps adding sources.
Caveats
- Survey respondents skew toward enthusiasts, so newer technologies (Rust, Elixir, Svelte) likely have supply overstated and scarcity understated.
- Job postings tag languages and cloud platforms more consistently than frameworks; compare frameworks with each other.
- Broad technologies such as Python and AWS appear in many non-engineering postings, which inflates their demand.