Quick wins for a faster PC:
Fix the driver behind crashes, sound loss and screen glitchesFind Drivers →Repair Windows errors before they cause bigger problemsFix Now →Scan for outdated or missing drivers - takes under a minuteDriver Scan →For U.S. readers, the programming job market in 2026 combines a positive long-run outlook with a difficult entry point. The newest Bureau of Labor Statistics (BLS) projections show software-developer employment rising through 2035, but LinkedIn’s February 2026 hiring analysis found that entry-level software-engineering hiring had not recovered by the end of 2025. AI is changing the tasks and skills employers want, but the evidence does not show that it has made software careers obsolete. Whether the field is worth investing in depends on which role you mean, what your training costs, and how you would realistically get your first job.
Defining the occupations before reading the numbers
“Programming” is a loose label. BLS tracks separate occupations in its Occupational Outlook Handbook and Employment Projections, and they are moving in different directions. Software developers analyze users’ needs and design, develop, and maintain software, and they may also work across testing and maintenance. Computer programmers primarily write, modify, and test code and scripts. A forecast for one title should not be read as a forecast for every coding job.
As an Amazon Associate I earn from qualifying purchases.
| Occupation (BLS category) | Projected employment change, 2025–35 | Employment, 2025 to 2035 | Average annual openings | Median annual wage, May 2025 |
|---|---|---|---|---|
| Software developers | +10.2% | 1,717,800 to 1,892,600 (net increase of 174,700) | 95,300 | $135,980 |
| Computer programmers | -7.3% | 110,800 to 102,700 | 4,400 | Not stated in the BLS figures cited here |
| Software developers, QA analysts, and testers (combined group) | +10% (all occupations: +3%) | Not stated in the BLS figures cited here | About 106,100 | Not stated in the BLS figures cited here |
Openings include both new positions created by growth and positions opened by people leaving jobs, so they describe how many roles become available, not how easily a newcomer can fill one. The combined row covers a different group from the software-developer row and should not be added to it or used in its place.
What the long-run projections do and do not say
For software developers, the BLS 2025–35 projections point to continued growth, and they are the current baseline for U.S. planning. The contrast with computer programmers is the most useful signal in the table. The projections show that the two categories are on different paths, so anyone aiming at software work should look at the developer category directly and treat the narrower programmer category as a separate question. The data do not explain why the programmer category is shrinking.
#1 Best Overall
- Careercup, Easy To Read
- Condition : Good
- Compact for travelling
The combined developer, QA, and tester group is also projected to grow faster than the all-occupation average. Because it blends several roles, cite it only as a combined figure.
Pay: a national median, not a starting offer
BLS reports a median annual wage of $135,980 for software developers in May 2025. Half of workers in the occupation earned more than that figure and half earned less, so it describes the middle of the national distribution. It does not describe what a graduate or career changer will be offered. Local wage levels, employer size, and industry all shift pay, and the BLS estimate is national.
Rank #2
Entry-level hiring: the hard part in 2026
Long-run projections do not measure how many openings are realistically open to newcomers right now. The most specific recent view is LinkedIn’s February 2026 U.S. report, which draws on LinkedIn Economic Graph platform data. Its hiring observations run through the end of 2025.
How hiring slowed
According to the report, software-engineering hiring slowed sharply from mid-2022 to late 2023, and entry-level hiring had not rebounded by the end of 2025. LinkedIn found that software-engineering hiring broadly tracked national and technology-sector hiring. It also concluded that the evidence is not enough to show that AI caused the entry-level slowdown, and that AI’s role in slowing recovery remains unclear. The accurate description is a difficult and unsettled entry market without one proven cause.
Where software-engineering jobs are being hired
From 2021 through 2025, LinkedIn attributed 35% of software-engineering hiring to the technology sector, 21% to professional services, 13% to manufacturing, and 12% to financial services. The report also shows rising hiring shares for AI Engineer and Machine Learning Engineer roles, while general software roles remain significant. For a job seeker, the practical lesson is to search beyond software companies. It does not mean every sector is hiring at the same pace.
What computer-science graduates are doing
LinkedIn reports that 55% of computer-science graduates in 2023 and 2024 started in roles outside software engineering, compared with 49% of graduates in 2016. On these figures, a computer-science degree is not a direct route to a software-engineering first job for most graduates.
AI: changing the work without a settled net effect
A February 2025 BLS Monthly Labor Review article by Christine Machovec, an economist in BLS’s Office of Employment and Unemployment Statistics, explains how generative AI can affect programming work, including developing, testing, and documenting code. The article states: “Programming is one of many work activities in which new LLMs and GenAI are well suited to augment worker efforts and increase productivity.”
Two effects that pull in opposite directions
The article’s central point is that the net employment effect on software developers is uncertain. Higher productivity can reduce the labor needed for each task. Lower software costs, broader demand for software, and the need to build and maintain AI systems can all support employment. BLS concluded that strong software needs were expected to support continued demand under its 2023–33 projection assumptions. Those projections have been superseded by the 2025–35 baseline used above, so use the article for its reasoning about these mechanisms rather than as the current growth figure.
Best Value
Education: the typical entry route
BLS lists a bachelor’s degree in computer and information technology or a related field as the typical preparation for software developers, QA analysts, and testers, and notes that some employers prefer a master’s degree. This describes the typical entry requirement in the BLS occupational profile. It is not a universal hiring rule, and it does not establish that a degree has positive financial value for every student.
Independent reader supportYour contribution helps us test, update, and keep practical guides available for everyone.Skills that appear in hiring data
LinkedIn’s 2021–25 analysis of software-engineering postings and hires points to a consistent set of skills:
- Common in job postings: software development, Java, JavaScript, Python, and AWS
- Prominent among hires: Python, AWS, SQL, JavaScript, and React.js
- Fast-growing: AI, Tailwind CSS, and data and cloud platforms
The pattern supports a practical approach: keep core programming skills and add the cloud, data, or AI skills that your target role requires. The same data does not identify one best language or one credential.
Is it still worth investing? A decision framework
A defensible answer is that it can be worth it, with a deliberate route and realistic expectations. The long-run projections and wage level are favorable for software development, the entry market is competitive, and the payoff depends on your own costs and starting point. The table shows how different starting points change the picture.
Quick Recap
| Your starting point | What the evidence supports | Main risk to plan for |
|---|---|---|
| Career changer with no technology work history | Long-run developer projections are positive, but entry-level software-engineering hiring had not rebounded by the end of 2025. | Competition for first roles, which makes demonstrated skill more important. |
| Already working in an adjacent technology role, such as QA or data | The combined developer, QA, and tester group is projected to grow faster than average. | The cited sources do not measure internal transitions into development roles separately. |
| Targeting AI or machine-learning engineering | LinkedIn reports rising hiring shares for AI Engineer and Machine Learning Engineer roles. | Titles and requirements vary by employer, so check current postings for the specific skills asked. |
| Targeting only the “computer programmer” title | BLS projects employment in that category to decline 7.3% over 2025–35. | A narrow title may shrink even when broader software work grows; a developer or adjacent role may be a better target. |
| Living outside the United States | The U.S. figures in this article do not apply directly. | Use your country’s labor statistics and local job postings instead. |
A low-cost way to test the decision
- Name one target title, such as software developer, QA engineer, or machine-learning engineer, and search current job postings in the city or region where you can realistically work.
- In a sample of those postings, record the required skills, experience level, and education, and note how many accept candidates without prior work experience.
- Price the route: tuition or course fees, time away from paid work, and the first-job pay you can verify in local postings.
- Build one project you can explain in detail, covering design choices, testing, maintenance, and how you worked with others. BLS’s description of the developer role includes these activities alongside writing code.
- Set a checkpoint before you commit to paid training. Decide in advance what result, such as interviews or offers within a set period, would tell you to continue or stop.
What these figures cannot tell you
- The BLS numbers are national U.S. estimates and projections. They are not live counts of open job postings.
- LinkedIn’s findings come from its platform data and are not a census of all U.S. graduates or employers. Its most recent hiring observations run to the end of 2025, so check current postings for conditions in 2026.
- No source here calculates the return on investment of a degree or course for an individual. That calculation requires your costs, prior experience, location, and realistic starting pay.
- The evidence does not cover countries outside the United States.
Product prices and availability are accurate as of the date/time indicated and are subject to change. Any price and availability information displayed on Amazon at the time of purchase will apply.
