<?xml version="1.0" encoding="UTF-8"?>
<rss version="2.0" xmlns:atom="http://www.w3.org/2005/Atom" xmlns:dc="http://purl.org/dc/elements/1.1/">
  <channel>
    <title>DEV Community: Global Software Companies</title>
    <description>The latest articles on DEV Community by Global Software Companies (global-software-companies).</description>
    <link>https://dev.to/global-software-companies</link>
    <image>
      <url>https://media2.dev.to/dynamic/image/width=90,height=90,fit=cover,gravity=auto,format=auto/https:%2F%2Fdev-to-uploads.s3.us-east-2.amazonaws.com%2Fuploads%2Forganization%2Fprofile_image%2F13047%2Feddcce53-95f2-4a7e-867c-69b65681f2c8.png</url>
      <title>DEV Community: Global Software Companies</title>
      <link>https://dev.to/global-software-companies</link>
    </image>
    <atom:link rel="self" type="application/rss+xml" href="https://dev.to/feed/global-software-companies"/>
    <language>en</language>
    <item>
      <title>Programming Languages and Tools Named in Entry-Level Software Job Listings, 2026</title>
      <dc:creator>Global Software Companies Team</dc:creator>
      <pubDate>Thu, 05 Nov 2026 12:49:58 +0000</pubDate>
      <link>https://dev.to/global-software-companies/programming-languages-and-tools-named-in-entry-level-software-job-listings-2026-1def</link>
      <guid>https://dev.to/global-software-companies/programming-languages-and-tools-named-in-entry-level-software-job-listings-2026-1def</guid>
      <description>&lt;p&gt;We analyzed 2,517 US software listings labeled “Entry level” on LinkedIn, January–August 2026. Python was named most often, but it frequently appeared as an example or an alternative rather than an individual requirement. 5% named an AI coding tool or assistant as a qualification.&lt;/p&gt;

&lt;p&gt;If you're starting out in software development, the list of languages in a job listing can look like an intimidating checklist: Python, Java, C++, JavaScript, SQL. Often, it's a menu.&lt;/p&gt;

&lt;p&gt;At Global Software Companies (GSC), we analyzed 2,517 US software listings labeled “Entry level” on LinkedIn between January and August 2026, counting the languages and tools each one named as a qualification, whether required, preferred or optional. The typical listing named two programming languages. When Python appeared among required qualifications, it was usually one choice among several or an example.&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Python was named as a qualification in 46% of listings, ahead of Java (32%), C++ (31%) and JavaScript (27%).&lt;/li&gt;
&lt;li&gt;Where Python appeared among required qualifications, 61% of listings gave it as an example or one of several accepted languages.&lt;/li&gt;
&lt;li&gt;C++ was named in 73% of the 656 embedded and hardware-related listings, against 16% of the rest.&lt;/li&gt;
&lt;li&gt;Only 131 listings (5%) named an AI coding tool or assistant as a qualification.&lt;/li&gt;
&lt;/ul&gt;

&lt;h2&gt;
  
  
  Which Programming Languages Do Entry-Level Software Listings Name?
&lt;/h2&gt;

&lt;p&gt;Python was named most often: in 1,156 of 2,517 listings (46%). Java (32%), C++ (31%) and JavaScript (27%) followed. 2,135 listings (85%) named at least one programming language as a qualification.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Feditor-api.globalsoftwarecompanies.com%2Fassets%2F9f1f4647-774d-4098-9ef2-5bd873432542" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Feditor-api.globalsoftwarecompanies.com%2Fassets%2F9f1f4647-774d-4098-9ef2-5bd873432542" alt="Programming languages named as a qualification in 2,517 US entry-level software listings on LinkedIn, January–August 2026, with the number listing each among required qualifications: Python 45.9% (1,156, 935 required); Java 32.4% (815, 694 required); C++ 30.8% (776, 638 required); JavaScript 26.8% (675, 549 required); C# 22.3% (562, 470 required); SQL 20.6% (518, 430 required); TypeScript 18.4% (463, 361 required); C 16.9% (425, 340 required); Go 9.3% (234, 180 required); Bash 5.9% (149, 94 required)." width="1280" height="720"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Java and C++ were close: 815 and 776 listings.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Language&lt;/th&gt;
&lt;th&gt;Named as a qualification&lt;/th&gt;
&lt;th&gt;Listed among required&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Python&lt;/td&gt;
&lt;td&gt;45.9%&lt;/td&gt;
&lt;td&gt;37.1%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Java&lt;/td&gt;
&lt;td&gt;32.4%&lt;/td&gt;
&lt;td&gt;27.6%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;C++&lt;/td&gt;
&lt;td&gt;30.8%&lt;/td&gt;
&lt;td&gt;25.3%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;JavaScript&lt;/td&gt;
&lt;td&gt;26.8%&lt;/td&gt;
&lt;td&gt;21.8%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;C#&lt;/td&gt;
&lt;td&gt;22.3%&lt;/td&gt;
&lt;td&gt;18.7%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;SQL&lt;/td&gt;
&lt;td&gt;20.6%&lt;/td&gt;
&lt;td&gt;17.1%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;TypeScript&lt;/td&gt;
&lt;td&gt;18.4%&lt;/td&gt;
&lt;td&gt;14.3%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;C&lt;/td&gt;
&lt;td&gt;16.9%&lt;/td&gt;
&lt;td&gt;13.5%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Go&lt;/td&gt;
&lt;td&gt;9.3%&lt;/td&gt;
&lt;td&gt;7.2%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Bash&lt;/td&gt;
&lt;td&gt;5.9%&lt;/td&gt;
&lt;td&gt;3.7%&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;&lt;em&gt;Shares of 2,517 listings. Rust (4.2%) and PowerShell (2.9%) came next. &lt;a href="https://html.spec.whatwg.org/multipage/introduction.html#background" rel="noopener noreferrer"&gt;HTML&lt;/a&gt;{citation} (10.1%) and &lt;a href="https://www.w3.org/Style/CSS/Overview.en.html" rel="noopener noreferrer"&gt;CSS&lt;/a&gt;{citation} (8.7%) are markup and styling languages, so they aren't counted as programming languages here. Counting identical descriptions once (2,186 unique listings), C++ edges ahead of Java, 32.8% to 31.8%.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;The chart's split matters more than the ranking.&lt;/p&gt;

&lt;h2&gt;
  
  
  Does Naming a Language Mean It's Required?
&lt;/h2&gt;

&lt;p&gt;Often not. Python was listed among required qualifications in 935 listings (37%), but in 571 of those (61%) it was given as one of several accepted languages or as an example. A required line that says “Python, Java or C++” is offering a choice, not asking for all three. Another 221 listings named Python only as preferred or optional.&lt;/p&gt;

&lt;p&gt;Python isn't unusual. Of the ten languages named most often, eight followed the same pattern. In most listings that put the language among required qualifications, it was one option among several, or an example. The exceptions were SQL and Bash, which were more often listed as requirements in their own right.&lt;/p&gt;

&lt;h2&gt;
  
  
  Why Do C and C++ Rank So High?
&lt;/h2&gt;

&lt;p&gt;Largely because of the kind of work. 656 listings (26%) mentioned embedded or hardware-related work: firmware, drivers, real-time systems or hardware integration. C++ was named in 73% of those listings, against 16% of the other 1,861; C in 46%, against 7%.&lt;/p&gt;

&lt;p&gt;&lt;a href="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Feditor-api.globalsoftwarecompanies.com%2Fassets%2Fc4d9cd07-e801-4f3e-b3b3-33eccf91fccd" class="article-body-image-wrapper"&gt;&lt;img src="https://media2.dev.to/dynamic/image/width=800%2Cheight=%2Cfit=scale-down%2Cgravity=auto%2Cformat=auto/https%3A%2F%2Feditor-api.globalsoftwarecompanies.com%2Fassets%2Fc4d9cd07-e801-4f3e-b3b3-33eccf91fccd" alt="Share of US entry-level software listings naming each language as a qualification, January–August 2026, in 656 listings mentioning embedded or hardware work versus the other 1,861: C++ 72.6% versus 16.1%; Python 52.9% versus 43.5%; C 46.2% versus 6.6%; Java 21.5% versus 36.2%; C# 14.8% versus 25.0%; JavaScript 10.4% versus 32.6%." width="1280" height="720"&gt;&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;Outside embedded work, the list looks different. Python (43%), Java (36%) and JavaScript (33%) led, and C++ dropped to seventh. Which languages matter depends on what kind of software you want to build.&lt;/p&gt;

&lt;h2&gt;
  
  
  Which Frameworks, Cloud Platforms and Databases Do Listings Name?
&lt;/h2&gt;

&lt;p&gt;React was the framework named most often (19%), Amazon Web Services (AWS) the cloud platform (22%) and Microsoft SQL Server the database (10%). AWS was named about 1.6 times as often as Microsoft Azure.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Type&lt;/th&gt;
&lt;th&gt;Name&lt;/th&gt;
&lt;th&gt;Named as a qualification&lt;/th&gt;
&lt;th&gt;Listed among required&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Framework&lt;/td&gt;
&lt;td&gt;React&lt;/td&gt;
&lt;td&gt;19.0%&lt;/td&gt;
&lt;td&gt;12.6%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Framework&lt;/td&gt;
&lt;td&gt;.NET (incl. .NET Core, ASP.NET)&lt;/td&gt;
&lt;td&gt;14.2%&lt;/td&gt;
&lt;td&gt;11.6%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Framework&lt;/td&gt;
&lt;td&gt;Angular&lt;/td&gt;
&lt;td&gt;10.9%&lt;/td&gt;
&lt;td&gt;7.6%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cloud&lt;/td&gt;
&lt;td&gt;Amazon Web Services&lt;/td&gt;
&lt;td&gt;22.1%&lt;/td&gt;
&lt;td&gt;10.5%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cloud&lt;/td&gt;
&lt;td&gt;Microsoft Azure&lt;/td&gt;
&lt;td&gt;13.6%&lt;/td&gt;
&lt;td&gt;5.3%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cloud&lt;/td&gt;
&lt;td&gt;Google Cloud&lt;/td&gt;
&lt;td&gt;9.1%&lt;/td&gt;
&lt;td&gt;4.5%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Database&lt;/td&gt;
&lt;td&gt;Microsoft SQL Server&lt;/td&gt;
&lt;td&gt;9.7%&lt;/td&gt;
&lt;td&gt;8.2%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Database&lt;/td&gt;
&lt;td&gt;PostgreSQL&lt;/td&gt;
&lt;td&gt;9.1%&lt;/td&gt;
&lt;td&gt;5.8%&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Database&lt;/td&gt;
&lt;td&gt;MySQL&lt;/td&gt;
&lt;td&gt;5.0%&lt;/td&gt;
&lt;td&gt;3.6%&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;&lt;em&gt;Shares of 2,517 listings. Counting identical descriptions once, PostgreSQL was ahead of SQL Server (9.1% to 6.8%), so the two are close rather than ranked.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;Cloud skills were mostly a plus rather than a requirement: for each of the three platforms, more than half of the listings that named it did so only as preferred or optional. At least one of the three main cloud platforms was named in 657 listings (26%).&lt;/p&gt;

&lt;h2&gt;
  
  
  Which AI Coding Tools Do Entry-Level Listings Name?
&lt;/h2&gt;

&lt;p&gt;This was one of the more surprising aspects  of the data. Only 131 of 2,517 listings (5%) named at least one AI coding tool or AI assistant as a qualification, and 77 (3%) listed one among required qualifications. In 68 of those 77, the tool was one of several options or an example.&lt;/p&gt;

&lt;p&gt;Among dedicated coding tools, Anthropic's Claude Code (67 listings) and Cursor (49) were named most often, followed by OpenAI Codex (41) and GitHub Copilot (37). Among general-purpose assistants, Claude was named in 40 listings and ChatGPT in 19.&lt;/p&gt;

&lt;p&gt;Most of those counts are small, and a few come from a handful of employers.&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;AI tool&lt;/th&gt;
&lt;th&gt;Type&lt;/th&gt;
&lt;th&gt;Listings&lt;/th&gt;
&lt;th&gt;Listed among required&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Claude Code&lt;/td&gt;
&lt;td&gt;Coding tool&lt;/td&gt;
&lt;td&gt;67&lt;/td&gt;
&lt;td&gt;48&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Cursor&lt;/td&gt;
&lt;td&gt;Coding tool&lt;/td&gt;
&lt;td&gt;49&lt;/td&gt;
&lt;td&gt;38&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;OpenAI Codex&lt;/td&gt;
&lt;td&gt;Coding tool&lt;/td&gt;
&lt;td&gt;41&lt;/td&gt;
&lt;td&gt;17&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Claude&lt;/td&gt;
&lt;td&gt;General-purpose assistant&lt;/td&gt;
&lt;td&gt;40&lt;/td&gt;
&lt;td&gt;17&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;GitHub Copilot&lt;/td&gt;
&lt;td&gt;Coding tool&lt;/td&gt;
&lt;td&gt;37&lt;/td&gt;
&lt;td&gt;17&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;ChatGPT&lt;/td&gt;
&lt;td&gt;General-purpose assistant&lt;/td&gt;
&lt;td&gt;19&lt;/td&gt;
&lt;td&gt;8&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Copilot (product not specified)&lt;/td&gt;
&lt;td&gt;Not specified&lt;/td&gt;
&lt;td&gt;17&lt;/td&gt;
&lt;td&gt;11&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Microsoft Copilot&lt;/td&gt;
&lt;td&gt;General-purpose assistant&lt;/td&gt;
&lt;td&gt;12&lt;/td&gt;
&lt;td&gt;3&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Gemini&lt;/td&gt;
&lt;td&gt;General-purpose assistant&lt;/td&gt;
&lt;td&gt;8&lt;/td&gt;
&lt;td&gt;2&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;&lt;strong&gt;Any AI coding tool or assistant&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;—&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;131&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;77&lt;/strong&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;&lt;em&gt;A listing can name several tools; 11 listings named one of 7 other tools, each in fewer than 5 listings. Some counts come from few employers: Microsoft Copilot's 12 listings came from 3 organizations, and OpenAI Codex's 41 from 23. Counting identical descriptions once, GitHub Copilot moves ahead of Codex (37 to 30).&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;Even fewer asked applicants to check what those tools produce.&lt;/p&gt;

&lt;h2&gt;
  
  
  Do Listings Ask Applicants to Check AI-Generated Code?
&lt;/h2&gt;

&lt;p&gt;Despite this being a growing aspect of university courses only 70 listings (3%) explicitly mentioned checking or reviewing output from AI coding tools: 50 as a part of the job, 26 as a required qualification and 9 as a preferred one (a listing can do more than one).&lt;/p&gt;

&lt;p&gt;More listings mentioned AI in the work itself. 397 (16%) mentioned using AI tools, named or not, to help write, test or review code, as part of the job or as a qualification.&lt;/p&gt;

&lt;p&gt;These figures describe what listings say, not what employers check in practice. &lt;a href="https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-218.pdf#page=23" rel="noopener noreferrer"&gt;Testing and code-review&lt;/a&gt;{citation} duties aren't an explicit requirement to check AI output, and silence on checking doesn't make unchecked code acceptable.&lt;/p&gt;

&lt;h2&gt;
  
  
  How Should You Read the Languages in a Listing?
&lt;/h2&gt;

&lt;p&gt;GSC's advice for reading these listings:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Check whether a language list offers alternatives or requires several languages. A required line often names options rather than asking for all of them.&lt;/li&gt;
&lt;li&gt;Decide what kind of software you want to build, then check listings for that kind of role. Embedded and hardware listings named C and C++ far more often than others.&lt;/li&gt;
&lt;li&gt;Check whether cloud experience is required or preferred; in this dataset, each of the three main platforms was more often preferred or optional.&lt;/li&gt;
&lt;/ul&gt;

&lt;p&gt;To dig further, download &lt;a href="https://www.globalsoftwarecompanies.com/research-data/#software-job-listings-dataset" rel="noopener noreferrer"&gt;the dataset&lt;/a&gt; and filter it for the roles and places you're interested in, or start from our &lt;a href="https://www.kaggle.com/code/globalsoftwarecompan/entry-level-software-job-listings-get-started" rel="noopener noreferrer"&gt;worked notebook on Kaggle&lt;/a&gt;. For a wider view of the occupation, see the Bureau of Labor Statistics' &lt;a href="https://www.bls.gov/ooh/computer-and-information-technology/software-developers.htm" rel="noopener noreferrer"&gt;profile of software developers&lt;/a&gt;{citation} and the Department of Labor's &lt;a href="https://www.onetonline.org/link/details/15-1252.00" rel="noopener noreferrer"&gt;O*NET list of their software skills&lt;/a&gt;{citation}, which flags the technologies most frequently included in employer job postings.&lt;/p&gt;

&lt;h2&gt;
  
  
  Methods and Limits
&lt;/h2&gt;

&lt;p&gt;GSC analyzed 2,517 US software listings labeled “Entry level” on LinkedIn and posted January–August 2026, from its dataset &lt;em&gt;US entry-labeled software job listings&lt;/em&gt; (version 2026-10-12b): the listings in &lt;a href="https://www.globalsoftwarecompanies.com/entry-level-software-jobs-experience/" rel="noopener noreferrer"&gt;our analysis of experience requirements&lt;/a&gt; that asked for 2 years' experience or less, or stated no minimum, excluding internships and specialist AI roles. &lt;/p&gt;

&lt;p&gt;We classified each listing with AI assistance under written rules, a GSC researcher checked about 1 in 5 of the classifications by hand, and the AI tool names were checked against the listing text. “Named as a qualification” covers required, preferred and optional wording; “listed among required” includes lists of alternatives. Duties and the employer's own technology weren't counted.&lt;/p&gt;

&lt;p&gt;These are selected listings, not a representative sample of US jobs, and 78% were posted in July–August, so we make no trend claims. A listing that names a tool doesn't say how the job uses it. Every figure here can be recomputed from the download. &lt;a href="https://www.globalsoftwarecompanies.com/research-data/#methods" rel="noopener noreferrer"&gt;Read the full methods&lt;/a&gt;.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;&lt;a href="https://www.globalsoftwarecompanies.com/research-data/#software-job-listings-dataset" rel="noopener noreferrer"&gt;Download the data&lt;/a&gt;&lt;/strong&gt; and &lt;strong&gt;&lt;a href="https://www.globalsoftwarecompanies.com/research-data/#methods" rel="noopener noreferrer"&gt;read how we did it&lt;/a&gt;&lt;/strong&gt;. Related: &lt;a href="https://www.globalsoftwarecompanies.com/entry-level-software-jobs-experience/" rel="noopener noreferrer"&gt;We analyzed 5,625 “entry-level” software jobs&lt;/a&gt; and &lt;a href="https://www.globalsoftwarecompanies.com/do-you-need-a-degree-entry-level-software-jobs/" rel="noopener noreferrer"&gt;Do you need a degree for an entry-level software job?&lt;/a&gt;&lt;/p&gt;

&lt;p&gt;&lt;em&gt;Last updated 4 November 2026. Corrections: &lt;a href="mailto:hello@globalsoftwarecompanies.com"&gt;hello@globalsoftwarecompanies.com&lt;/a&gt;.&lt;/em&gt;&lt;/p&gt;

&lt;p&gt;This article draws on GSC’s analysis and the outside sources linked in the text:&lt;/p&gt;

&lt;ul&gt;
&lt;li&gt;Global Software Companies, &lt;a href="https://www.globalsoftwarecompanies.com/research-data/#software-job-listings-dataset" rel="noopener noreferrer"&gt;&lt;em&gt;US entry-labeled software job listings&lt;/em&gt;&lt;/a&gt;, version 2026-10-12b, CC BY 4.0. Definitions, selection rules and limitations: &lt;a href="https://www.globalsoftwarecompanies.com/research-data/#methods" rel="noopener noreferrer"&gt;methods&lt;/a&gt;.&lt;/li&gt;
&lt;/ul&gt;

&lt;h3&gt;
  
  
  Q: Which programming language do entry-level software job listings name most often?
&lt;/h3&gt;

&lt;p&gt;Python: 1,156 of 2,517 listings (46%) named it as a qualification, and 935 (37%) listed it among required qualifications.&lt;/p&gt;

&lt;h3&gt;
  
  
  Q: If a listing names a language, is it required?
&lt;/h3&gt;

&lt;p&gt;Not always. In 571 of the 935 listings that placed Python among required qualifications, it was an example or one of several accepted languages. Read the wording around the list.&lt;/p&gt;

&lt;h3&gt;
  
  
  Q: Which language should I learn for an entry-level software job?
&lt;/h3&gt;

&lt;p&gt;It depends on the work. Outside embedded and hardware roles, Python, Java and JavaScript were named most often; in embedded and hardware listings, C++ was named in 73% and C in 46%.&lt;/p&gt;

&lt;h3&gt;
  
  
  Q: Do entry-level listings ask for AI coding tools?
&lt;/h3&gt;

&lt;p&gt;Rarely: 131 of 2,517 listings (5%) named one as a qualification, and no single tool appeared in more than 67. Of the 77 that listed one among required qualifications, 68 gave it as an example or an option.&lt;/p&gt;

&lt;h3&gt;
  
  
  Q: Can I use this data?
&lt;/h3&gt;

&lt;p&gt;Yes. The GSC jobs dataset is free to download under &lt;a href="https://creativecommons.org/licenses/by/4.0/" rel="noopener noreferrer"&gt;CC BY 4.0&lt;/a&gt; from our &lt;a href="https://www.globalsoftwarecompanies.com/research-data/#software-job-listings-dataset" rel="noopener noreferrer"&gt;research data page&lt;/a&gt;; credit Global Software Companies.&lt;/p&gt;

&lt;h3&gt;
  
  
  Q: How often is this page updated?
&lt;/h3&gt;

&lt;p&gt;When GSC releases a new version of the dataset. The update history below lists every update.&lt;/p&gt;

&lt;h2&gt;
  
  
  Other sources:
&lt;/h2&gt;

&lt;ul&gt;
&lt;li&gt;US Bureau of Labor Statistics, “&lt;a href="https://www.bls.gov/ooh/computer-and-information-technology/software-developers.htm" rel="noopener noreferrer"&gt;Software Developers, Quality Assurance Analysts, and Testers&lt;/a&gt;{citation}”, &lt;em&gt;Occupational Outlook Handbook&lt;/em&gt;.&lt;/li&gt;
&lt;li&gt;O*NET OnLine, “&lt;a href="https://www.onetonline.org/link/details/15-1252.00" rel="noopener noreferrer"&gt;15-1252.00 — Software Developers&lt;/a&gt;{citation}”, US Department of Labor.&lt;/li&gt;
&lt;li&gt;WHATWG, “&lt;a href="https://html.spec.whatwg.org/multipage/introduction.html#background" rel="noopener noreferrer"&gt;HTML Standard: Background&lt;/a&gt;{citation}”: HTML as a markup language.&lt;/li&gt;
&lt;li&gt;World Wide Web Consortium (W3C), “&lt;a href="https://www.w3.org/Style/CSS/Overview.en.html" rel="noopener noreferrer"&gt;Cascading Style Sheets&lt;/a&gt;{citation}”: CSS and document styling.&lt;/li&gt;
&lt;li&gt;National Institute of Standards and Technology (NIST), “&lt;a href="https://nvlpubs.nist.gov/nistpubs/SpecialPublications/NIST.SP.800-218.pdf#page=23" rel="noopener noreferrer"&gt;Secure Software Development Framework, Version 1.1&lt;/a&gt;{citation}”, SP 800-218, February 2022, practices PW.7–PW.8: background on code review and testing.&lt;/li&gt;
&lt;li&gt;Creative Commons, “&lt;a href="https://creativecommons.org/licenses/by/4.0/" rel="noopener noreferrer"&gt;Attribution 4.0 International (CC BY 4.0)&lt;/a&gt;{citation}”: dataset reuse licence terms.&lt;/li&gt;
&lt;/ul&gt;

</description>
      <category>career</category>
      <category>programming</category>
      <category>softwaredevelopment</category>
    </item>
    <item>
      <title>4,552 Developers Called Themselves AI Developers in 2018. In 2025, Only 320 Did</title>
      <dc:creator>Global Software Companies Team</dc:creator>
      <pubDate>Thu, 23 Apr 2026 09:47:28 +0000</pubDate>
      <link>https://dev.to/global-software-companies/4552-developers-called-themselves-ai-developers-in-2018-in-2025-only-320-did-g1l</link>
      <guid>https://dev.to/global-software-companies/4552-developers-called-themselves-ai-developers-in-2018-in-2025-only-320-did-g1l</guid>
      <description>&lt;p&gt;In the 2018 Stack Overflow Developer Survey, 4,552 developers identified themselves as AI developers. By 2025 that number was 320 -- a 93% collapse over seven years. The counterintuitive part: the ones still claiming the title are earning more than ever. Median AI developer salary jumped 32% in a single year, the largest single-year wage move across any specialization we tracked from 2024 to 2025.&lt;/p&gt;

&lt;p&gt;A category that's simultaneously disappearing and commanding premium pay is an unusual pattern worth understanding.&lt;/p&gt;




&lt;h2&gt;
  
  
  The AI Developer Arc, 2018-2025
&lt;/h2&gt;

&lt;p&gt;The year-by-year picture from the Stack Overflow survey series:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Year&lt;/th&gt;
&lt;th&gt;Respondents&lt;/th&gt;
&lt;th&gt;Median Salary (USD)&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;2018&lt;/td&gt;
&lt;td&gt;4,552&lt;/td&gt;
&lt;td&gt;$61,194&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2019&lt;/td&gt;
&lt;td&gt;5,610&lt;/td&gt;
&lt;td&gt;$60,000&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2020&lt;/td&gt;
&lt;td&gt;3,764&lt;/td&gt;
&lt;td&gt;$55,217&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2021&lt;/td&gt;
&lt;td&gt;5,434&lt;/td&gt;
&lt;td&gt;$54,049&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2022&lt;/td&gt;
&lt;td&gt;3,710&lt;/td&gt;
&lt;td&gt;$64,500&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2023&lt;/td&gt;
&lt;td&gt;4,117&lt;/td&gt;
&lt;td&gt;$74,963&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2024&lt;/td&gt;
&lt;td&gt;2,433&lt;/td&gt;
&lt;td&gt;$64,444&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;2025&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;320&lt;/strong&gt;&lt;/td&gt;
&lt;td&gt;&lt;strong&gt;$85,230&lt;/strong&gt;&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;Three things stand out.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The collapse accelerated dramatically in 2024-2025.&lt;/strong&gt; From 2018 through 2024, the AI developer count fluctuated between roughly 2,400 and 5,600 -- noisy but bounded. Between 2024 and 2025 the category lost nearly 87% of its self-identified population in a single year.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Pay trajectory was flat-ish until late.&lt;/strong&gt; Through 2021 the median drifted downward from the low-$60s to the mid-$50s -- effectively declining in real terms. 2022 saw the first meaningful step up. 2025 produced the biggest single-year jump in the series: +32% year-over-year to $85,230.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The small-sample caveat matters.&lt;/strong&gt; 320 respondents is below the threshold we'd normally use for a representative global median in a developer specialization. Treat the 2025 figure as directionally informative, not definitive -- the trend is real, but the specific dollar value may shift when the 2026 survey lands.&lt;/p&gt;




&lt;h2&gt;
  
  
  Meanwhile, DevOps Went the Other Way
&lt;/h2&gt;

&lt;p&gt;In 2024, DevOps Services had 586 Stack Overflow respondents globally -- too few to report as a representative median. In 2025, DevOps had 13,010 respondents. That's a 22x increase in a single year.&lt;/p&gt;

&lt;p&gt;Unlike AI, DevOps now has a sample large enough to support a market-representative median: $84,691. Here's the 2025 ranking across specializations with 2,000 or more respondents:&lt;/p&gt;

&lt;div class="table-wrapper-paragraph"&gt;&lt;table&gt;
&lt;thead&gt;
&lt;tr&gt;
&lt;th&gt;Specialization&lt;/th&gt;
&lt;th&gt;2025 Median&lt;/th&gt;
&lt;th&gt;Respondents&lt;/th&gt;
&lt;/tr&gt;
&lt;/thead&gt;
&lt;tbody&gt;
&lt;tr&gt;
&lt;td&gt;Cloud Consulting&lt;/td&gt;
&lt;td&gt;$90,000&lt;/td&gt;
&lt;td&gt;12,210&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;DevOps Services&lt;/td&gt;
&lt;td&gt;$84,691&lt;/td&gt;
&lt;td&gt;13,010&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Mobile App Development&lt;/td&gt;
&lt;td&gt;$81,000&lt;/td&gt;
&lt;td&gt;3,605&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Web Development&lt;/td&gt;
&lt;td&gt;$80,254&lt;/td&gt;
&lt;td&gt;17,989&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;IoT Development&lt;/td&gt;
&lt;td&gt;$80,050&lt;/td&gt;
&lt;td&gt;16,529&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;API Development&lt;/td&gt;
&lt;td&gt;$80,000&lt;/td&gt;
&lt;td&gt;9,924&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;Custom Software Development&lt;/td&gt;
&lt;td&gt;$75,943&lt;/td&gt;
&lt;td&gt;12,625&lt;/td&gt;
&lt;/tr&gt;
&lt;tr&gt;
&lt;td&gt;E-Commerce Development&lt;/td&gt;
&lt;td&gt;$65,000&lt;/td&gt;
&lt;td&gt;3,823&lt;/td&gt;
&lt;/tr&gt;
&lt;/tbody&gt;
&lt;/table&gt;&lt;/div&gt;

&lt;p&gt;AI Development (n=320) and Cybersecurity (n=387) fall below the 2,000-respondent threshold and are excluded from this ranking. Blockchain Development has no 2025 figures in the survey series.&lt;/p&gt;

&lt;p&gt;Two specializations, opposite trajectories. DevOps consolidated into a clearly-named category that engineers increasingly identify with. AI dissolved.&lt;/p&gt;




&lt;h2&gt;
  
  
  Why One Category Grew and One Dissolved
&lt;/h2&gt;

&lt;p&gt;The data doesn't explain the &lt;em&gt;cause&lt;/em&gt; of the collapse. But there are a few hypotheses worth considering.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;AI work has become ambient.&lt;/strong&gt; In 2020, "I'm an AI developer" meant you worked on ML pipelines, model training, or MLOps -- a discrete set of responsibilities that looked different from standard backend or data work. In 2025, every backend developer integrates LLMs, every data engineer builds ML features, and most frontend developers use &lt;a href="https://github.com/features/copilot" rel="noopener noreferrer"&gt;Copilot&lt;/a&gt; or &lt;a href="https://cursor.com/" rel="noopener noreferrer"&gt;Cursor&lt;/a&gt; daily. &lt;a href="https://survey.stackoverflow.co/2024/ai" rel="noopener noreferrer"&gt;Stack Overflow's 2024 AI survey&lt;/a&gt; reported roughly 76% of professional developers using or planning to use AI tools in their development process. The work has spread into every other category, and the specialization label has lost much of its distinguishing power.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;DevOps didn't dissolve because the tooling stayed specialized.&lt;/strong&gt; Platform engineering, Kubernetes operations, CI/CD pipelines, and cloud cost optimization remain a discrete career track with specialized tools and identifiable responsibilities. A developer can "use" DevOps the way they now "use" AI, but the core work and the specialist title haven't collapsed into every other role.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Survey taxonomy may have shifted.&lt;/strong&gt; Stack Overflow periodically refines its category definitions. If the 2025 survey restructured how AI developers self-identify -- for example, folding "AI developer" into a broader "developer using AI tools" bucket -- that alone could drive a significant share of the count drop. We can't cleanly separate category redefinition from genuine dissolution in the data as published.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The remaining AI developers look more senior.&lt;/strong&gt; When a specialization disappears into general developer roles but a core group keeps the title, that core tends to skew toward specialists still doing distinct AI-specific work -- research roles, core ML engineering, foundation model training. A smaller, more senior pool commanding premium pay is consistent with the jump to $85,230 on just 320 respondents.&lt;/p&gt;




&lt;h2&gt;
  
  
  What This Means If You're a Developer
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;If you've been holding out for "AI developer" as a distinct career title, it's a shrinking pond.&lt;/strong&gt; Not necessarily a worse one -- pay is rising -- but the number of people calling themselves AI developers is falling fast.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The generalist developer who uses AI well is the new default.&lt;/strong&gt; The skill premium hasn't disappeared, but it's being absorbed into baseline expectations for all software engineers rather than concentrating in a specialist title.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;If you want the specialist track, expect it narrower and more competitive.&lt;/strong&gt; The 320 respondents in 2025 likely skew toward senior ML engineers, researchers, and core AI infrastructure roles. Pay reflects that. So does the competition.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;DevOps reads as a category where scale and specialization coexist.&lt;/strong&gt; 13,010 respondents is a meaningful population, and DevOps sits near the top of the pay ranking. If you're choosing between deepening into AI specialization versus platform engineering, DevOps is the larger addressable role market right now.&lt;/p&gt;




&lt;h2&gt;
  
  
  What This Doesn't Tell You
&lt;/h2&gt;

&lt;p&gt;&lt;strong&gt;It doesn't isolate the cause of the collapse.&lt;/strong&gt; Category dissolution and survey taxonomy changes are both plausible, and both probably contribute. The data shows the effect but not the cleanly-separable cause.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;It doesn't predict what happens next.&lt;/strong&gt; If the 2026 survey shows AI developer count rebounding, the dissolution hypothesis weakens. If it stays below 500, the pattern is real. Worth watching.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;Salary data is self-reported and self-selected.&lt;/strong&gt; Stack Overflow's survey skews toward developers who choose to participate and who communicate in English. Global medians are directional, not authoritative.&lt;/p&gt;

&lt;p&gt;&lt;strong&gt;The small-sample caveat for AI in 2025 is a real limit.&lt;/strong&gt; An $85,230 median on n=320 has wide confidence bounds. The trend direction is defensible; the specific figure is not precise.&lt;/p&gt;




&lt;h2&gt;
  
  
  Methodology
&lt;/h2&gt;

&lt;p&gt;Salary and respondent-count data is from the 2018-2025 &lt;a href="https://survey.stackoverflow.co/" rel="noopener noreferrer"&gt;Stack Overflow Developer Survey&lt;/a&gt; editions. For the 2025 cross-specialization ranking, we included only categories with 2,000 or more respondents -- the standard threshold for reporting statistically representative global medians. AI Development (n=320), Cybersecurity (n=387), and Blockchain Development (no 2025 data) fall below this threshold and are excluded from the cross-specialization ranking.&lt;/p&gt;

&lt;p&gt;Historical AI Development figures (2018-2024) come from the same survey series. The 2024 figures were above the 2,000-threshold; the 2025 figure is reported here because the sample collapse itself is the story, not because the figure meets our standard inclusion threshold.&lt;/p&gt;

&lt;p&gt;All figures are global medians in USD. Survey respondents are self-selected and skew toward English-speaking markets. Treat these figures as directional benchmarks, not definitive compensation data.&lt;/p&gt;

</description>
      <category>career</category>
      <category>ai</category>
      <category>data</category>
      <category>discuss</category>
    </item>
  </channel>
</rss>
