Global median pay for a back-end developer is USD 79,742, while the same role in the United States reaches USD 175,000. That gap, more than a factor of two, is the dominant fact in developer compensation data and it widens rather than narrows with seniority. Engineering managers sit at USD 130,000 globally against USD 200,000 in the US. The figures below come from the most recent Stack Overflow Developer Survey, which drew 48,178 total respondents with 23,928 supplying usable salary data.
TL;DR
- Global median for a senior executive is USD 139,218 (Stack Overflow)
- Engineering manager global median is USD 130,000 (Stack Overflow)
- Cloud infrastructure engineer global median is USD 103,112.50 (Stack Overflow)
- AI and machine learning engineer global median is USD 89,427 (Stack Overflow)
- Back-end developer global median is USD 79,742 (Stack Overflow)
- Front-end developer global median is USD 62,015 (Stack Overflow)
- QA and test developer global median is USD 57,442.50 (Stack Overflow)
- US senior executive median is USD 225,000 (Stack Overflow)
- US engineering manager median is USD 200,000 (Stack Overflow)
- US AI and machine learning engineer median is USD 189,500 (Stack Overflow)
- 32.4% of respondents work fully remote (Stack Overflow)
- 45% of US respondents work fully remote (Stack Overflow)
- 48,178 developers responded in total (Stack Overflow)
- 23,928 provided salary data, a 48.8% response rate (Stack Overflow)
1. Global Median Pay by Role
The global table spans a factor of 2.4 from top to bottom among the roles reported. Senior executives lead at USD 139,218, engineering managers follow at USD 130,000, and the range descends through architect at USD 102,000 and product manager at USD 100,000 to QA and test developer at USD 57,442.50.
What stands out is where the management premium sits. The jump from architect at USD 102,000 to engineering manager at USD 130,000 is about 27%, which is a substantial reward for moving off the technical track. But the jump from engineering manager to senior executive is only about 7%, far smaller than most career narratives imply. Compensation in this dataset rises steeply once and then flattens, which suggests the meaningful financial decision for most developers is whether to enter management at all rather than how far up they climb afterward.
| Metric | Value | Source |
|---|---|---|
| Senior executive, global median | USD 139,218 | Stack Overflow |
| Engineering manager | USD 130,000 | Stack Overflow |
| Financial analyst or engineer | USD 103,757 | Stack Overflow |
| Cloud infrastructure engineer | USD 103,112.50 | Stack Overflow |
| Architect, software or solutions | USD 102,000 | Stack Overflow |
| Product manager | USD 100,000 | Stack Overflow |
| QA and test developer | USD 57,442.50 | Stack Overflow |
| Ratio, top role to bottom role | approx. 2.4x | Derived from Stack Overflow medians |
Source: Stack Overflow Developer Survey, work and salary section.
2. The United States Premium
The country gap dwarfs the role gap. Engineering managers earn USD 130,000 globally against USD 200,000 in the United States, and back-end developers earn USD 79,742 globally against USD 175,000 in the US, a ratio of about 2.2 for the individual contributor role.
The asymmetry is the point. For engineering managers the US premium is about 54%, while for back-end developers it is about 120%. In other words the American market pays a far larger premium for ordinary engineering labour than it does for management, which compresses the internal hierarchy. A US back-end developer at USD 175,000 earns 87.5% of what a US engineering manager earns, whereas globally the same comparison is 61%. Geography changes not just the level of pay but the shape of the ladder.
| Metric | Value | Source |
|---|---|---|
| US senior executive median | USD 225,000 | Stack Overflow |
| US engineering manager | USD 200,000 | Stack Overflow |
| US AI and machine learning engineer | USD 189,500 | Stack Overflow |
| US cloud infrastructure engineer | USD 189,000 | Stack Overflow |
| US back-end developer | USD 175,000 | Stack Overflow |
| US premium, engineering manager | approx. 54% | Derived from Stack Overflow medians |
| US premium, back-end developer | approx. 120% | Derived from Stack Overflow medians |
| US back-end pay as share of US manager pay | approx. 87.5% | Derived from Stack Overflow medians |
Broader labour context sits in our AI impact on jobs statistics. Source: Stack Overflow Developer Survey, work and salary section.
3. Where the AI Premium Actually Shows Up
The AI premium is real but geographically concentrated, and the global figure alone would mislead. AI and machine learning engineers earn a global median of USD 89,427, which is above back-end developers at USD 79,742 but below cloud infrastructure engineers at USD 103,112.50. In the United States the same role reaches USD 189,500, third overall.
That inversion between the global and US tables is the finding worth attention. Globally, AI engineering is a moderately well-paid specialisation that does not out-earn cloud infrastructure work. In the US it nearly matches it and sits within striking distance of management pay. The reasonable inference is that the bidding war for AI talent is concentrated in a handful of high-cost markets rather than distributed across the profession, and developers reading headline AI salary figures should check which market produced them before drawing career conclusions.
| Metric | Value | Source |
|---|---|---|
| AI and ML engineer, global median | USD 89,427 | Stack Overflow |
| AI and ML engineer, US median | USD 189,500 | Stack Overflow |
| US premium for the role | approx. 112% | Derived from Stack Overflow medians |
| Cloud infrastructure engineer, global | USD 103,112.50 | Stack Overflow |
| Cloud infrastructure engineer, US | USD 189,000 | Stack Overflow |
| Does AI out-earn cloud globally | no | Derived from Stack Overflow medians |
| Does AI out-earn cloud in the US | marginally, by USD 500 | Derived from Stack Overflow medians |
| AI engineer US rank among listed roles | 3rd | Stack Overflow |
Source: Stack Overflow Developer Survey, work and salary section.
4. The Front-End Discount
One of the most consistent findings in developer compensation surveys reappears here. Front-end developers earn a global median of USD 62,015 against USD 79,742 for back-end, a gap of about USD 17,727 or roughly 29%.
Only QA and test developers rank lower among common engineering roles, at USD 57,442.50. The usual explanations are supply-side, since front-end work has the shallowest entry path and therefore the largest candidate pool, and there is nothing in this dataset that contradicts that reading. What the data does show is that the discount is not marginal. A front-end developer would need to reach roughly the 60th percentile of their own role’s distribution to match the median back-end salary, which makes the specialisation choice financially consequential in a way that job titles alone obscure.
| Metric | Value | Source |
|---|---|---|
| Front-end developer, global median | USD 62,015 | Stack Overflow |
| Back-end developer, global median | USD 79,742 | Stack Overflow |
| Absolute gap | approx. USD 17,727 | Derived from Stack Overflow medians |
| Relative gap | approx. 29% | Derived from Stack Overflow medians |
| QA and test developer, global median | USD 57,442.50 | Stack Overflow |
| Data scientist, global median | USD 82,910 | Stack Overflow |
| DevOps engineer, global median | USD 87,011 | Stack Overflow |
| Lowest-paid listed role | QA and test developer | Stack Overflow |
Tooling context sits in our developer survey statistics. Source: Stack Overflow Developer Survey, work and salary section.
5. How Developers Actually Work
The work model split is more fragmented than the remote-versus-office framing suggests. 32.4% work fully remote, 17.2% are hybrid leaning remote, 19.9% are hybrid leaning in-person, 17.9% are fully in-person, and 12.6% report a flexible arrangement of their own choosing.
Adding the two hybrid categories gives 37.1%, making hybrid the single largest bucket ahead of fully remote. That is a meaningful correction to the common claim that developers have settled into remote work; what they have settled into is a spread with no majority position. The US figure of 45% fully remote, the highest among surveyed countries, shows how much national labour markets diverge on this, and it partly explains why the American pay premium persists despite remote work theoretically flattening geographic arbitrage.
| Metric | Value | Source |
|---|---|---|
| Fully remote | 32.4% | Stack Overflow |
| Hybrid, remote-leaning | 17.2% | Stack Overflow |
| Hybrid, in-person-leaning | 19.9% | Stack Overflow |
| Fully in-person | 17.9% | Stack Overflow |
| Flexible, own choice | 12.6% | Stack Overflow |
| Combined hybrid share | 37.1% | Derived from Stack Overflow figures |
| US fully remote | 45% | Stack Overflow |
| Country with highest remote share | United States | Stack Overflow |
Deeper coverage sits in our remote work statistics. Source: Stack Overflow Developer Survey, work and salary section.
6. What the Sample Represents
Sample construction limits how far these numbers travel, and the survey is unusually transparent about it. Of 48,178 total respondents, 23,928 provided usable salary data, a 48.8% response rate on that question, with 69.8% employed and 13.9% self-employed.
Two caveats follow. Half the respondents declined to answer the salary question, and non-response on pay is rarely random, so the reported medians could shift in either direction if the missing half were included. More fundamentally, this is a self-selected sample of people who use Stack Overflow and choose to complete a long survey, which will skew toward certain seniority levels, languages, and regions. The dataset is large and internally consistent enough to compare roles against each other with confidence; treating any single figure as the population median for a profession of tens of millions is a different claim entirely.
| Metric | Value | Source |
|---|---|---|
| Total survey respondents | 48,178 | Stack Overflow |
| Respondents providing salary data | 23,928 | Stack Overflow |
| Salary question response rate | 48.8% | Stack Overflow |
| Employed | 69.8% | Stack Overflow |
| Self-employed | 13.9% | Stack Overflow |
| Sample type | self-selected, not random | Stack Overflow |
| Reliable use | comparing roles within the sample | Derived |
| Unreliable use | population-level absolute medians | Derived |
Employment volatility context sits in our tech layoffs statistics. Source: Independent analysis of the survey salary tables.
Summary: Developer Pay by the Numbers
| Metric | Value | Source |
|---|---|---|
| Senior executive, global | USD 139,218 | Stack Overflow |
| Engineering manager, global | USD 130,000 | Stack Overflow |
| Financial analyst or engineer, global | USD 103,757 | Stack Overflow |
| Cloud infrastructure engineer, global | USD 103,112.50 | Stack Overflow |
| Architect, global | USD 102,000 | Stack Overflow |
| Product manager, global | USD 100,000 | Stack Overflow |
| AI and ML engineer, global | USD 89,427 | Stack Overflow |
| DevOps engineer, global | USD 87,011 | Stack Overflow |
| Data scientist, global | USD 82,910 | Stack Overflow |
| Back-end developer, global | USD 79,742 | Stack Overflow |
| Front-end developer, global | USD 62,015 | Stack Overflow |
| QA and test developer, global | USD 57,442.50 | Stack Overflow |
| Senior executive, US | USD 225,000 | Stack Overflow |
| Engineering manager, US | USD 200,000 | Stack Overflow |
| AI and ML engineer, US | USD 189,500 | Stack Overflow |
| Cloud infrastructure engineer, US | USD 189,000 | Stack Overflow |
| Back-end developer, US | USD 175,000 | Stack Overflow |
| Fully remote | 32.4% | Stack Overflow |
| Combined hybrid | 37.1% | Derived |
| US fully remote | 45% | Stack Overflow |
| Total respondents | 48,178 | Stack Overflow |
| Salary responses | 23,928 | Stack Overflow |
Methodology and Sources
- All salary medians, work model percentages, and sample figures come from the Stack Overflow Developer Survey, work and salary section, covering 48,178 total respondents with 23,928 providing usable compensation data (Stack Overflow).
- Country-level breakdowns were cross-checked against independent analyses of the same published tables (TechRecruiting, Chartkite).
- Data watch: every figure here is self-reported by a self-selected sample, which is the central limitation and cannot be corrected after the fact. Roughly half of respondents did not answer the salary question, and non-response on compensation is unlikely to be random. Medians are reported in US dollars after conversion, so exchange rate movement between the survey period and reading affects non-US figures independently of any change in local pay. Global medians pool markets with vastly different costs of living and are not adjusted for purchasing power, which makes cross-country comparison of absolute values misleading even though the within-sample role comparisons hold up. Rows marked as derived are arithmetic on published medians. This survey is annual, so figures reflect a single point in time rather than a running average.
- Last updated: August 2, 2026. We update this roundup quarterly, and the next major refresh is expected when Stack Overflow publishes its next annual developer survey.