← Risk register SOC 15-1254 · reviewed 2026-08-11

Web Developers

70,190 US workers · median $92,650/yr · Tech

COOKED

The modal web developer builds and maintains sites and front-end features — HTML/CSS/JS, CMS theming, responsive layouts, form handling, API wiring, WordPress and Shopify customization — and that is precisely the work current models generate at usable quality in one pass. What persists is not the markup but the accountability: performance and accessibility audits that hold up under a lawsuit, security hardening, and the client conversation about what should be built. Note the split: agency/freelance site-builders are far more exposed than developers who own production systems, auth, payments, and uptime.

10-year outlook: Expect the front-end implementation tier to contract sharply over ten years while a smaller group of developers who own production systems, security, and client relationships absorbs the surviving work at higher pay.

US employment, 2021–2025-17.2%
84,82070,190 workers

Most of this decline happened after 2021 — it is not the pandemic dip.

The job count is not the verdict

This line is counted by the Bureau of Labor Statistics — the one figure on this page that isn't a judgement of ours. Headcount moves on demand, offshoring, demographics and the business cycle, and automation is one term among several, often not the loudest.

So a falling line is not evidence that AI did it, and a rising one is not evidence that it won't. Both happen in this register: some occupations resist automation and shrink anyway, others are highly automatable and keep growing.

BLS projection, 2024–2034

+7.5%

Percentage only. The projection counts a different population from the 70,190 above — it includes self-employed workers, which for this occupation is most of them, so the two headcounts are not comparable.

Exposed, but growing

AI can already do a lot of these tasks, and the BLS still expects +7.5% more of these jobs by 2034. Demand for the output is growing faster than the work is being automated away — the mechanism BLS gives for software developers, and the combination people most often misread as an error.

Different clocks. The score is what current AI could do to this work today. The projection is how many of these jobs will exist in 2034. Everything between the two — how fast employers actually adopt, whether demand grows in the meantime — is why they can point opposite ways without either being wrong.

~5,400 openings a year on average, including replacing people who leave.

One email if this score changes. Watch as many occupations as you like from the same address — no account, and nothing is sent on a schedule, only when a verdict actually moves.

Also known as — 24 job titles this covers

Titles reported by people doing this work, from the US Department of Labor's O*NET survey. If your job title is here, this page is about your work even though the name doesn't match.

WebmasterWeb EngineerWeb ProducerWeb ArchitectWeb DeveloperWeb ConsultantWeb ProgrammerWeb SpecialistWeb TechnicianWebsite ManagerjQuery DeveloperBack End EngineerNode.js DeveloperTableau DeveloperWeb Site DesignerWebmaster AnalystBack End DeveloperFront End EngineerInterface DesignerInternet ArchitectInternet DeveloperIntranet DeveloperReact.js DeveloperWeb Page Developer

Score — 24/100 resistance

Holding it up: trust premium (8/20). Weakest point: liability shield (1/20).

Five dimensions, 0–20 each, summed. Higher means more protected. The arithmetic is shown so you can check it: 6 + 2 + 1 + 8 + 7 = 24. · Scored 2026-08-11, and re-examined when evidence accumulates rather than on a schedule.

Task resistance 6/20

Core tasks are already automatable Scaffolding a responsive layout, wiring a REST call, writing a contact-form validator, restyling a WordPress theme, and fixing a CSS breakpoint are all single-prompt outputs now, and since those tasks dominate the modal ticket queue the score sits at 6 rather than 12 — the residue that isn't one-pass is debugging a flaky third-party embed and untangling a decade-old jQuery codebase.

Embodiment 2/20

Fully desk- and screen-based The job runs entirely on a laptop over Git and Figma; the 2 rather than 0 covers testing on real iPhones and tablets in a device lab and the occasional trip to a client office or on-site photo shoot for asset capture.

Liability shield 1/20

No licence, no signature requirement There is no licensure, no registration board, no continuing-education requirement — anyone can ship production HTML, and when a site leaks card data or fails WCAG the exposure lands on the business owner or the agency's E&O policy, not on your name.

Trust premium 8/20

Some relationship component Freelancers and agency devs keep clients through referral and the fact that you already know why their checkout is wired weird, which is real switching cost, but the deliverable is a working site the client can hand to any other shop — an 8 reflects relationships that make you preferred, not irreplaceable.

Judgment & accountability 7/20

Meaningful discretion You choose framework, state management, and CDN strategy and decide whether a hack ships before Friday, but scope, brand, and budget arrive from a designer, PM, or client — the 7 marks genuine technical discretion inside decisions someone else owns.

Confidence: high · reviewed 2026-08-11 · how scoring works

What this job involves — and which parts are yours

The verdict above describes this occupation as a whole. Almost nobody does the typical version of a job — tick what's actually in your week and see how your own mix sits.

AI already does these at usable quality

These still need a person

Active moats on the surviving side: trust, judgment

How to future-proof this job

Training paths for your skill gaps: MIT OpenCourseWare — systems analysis and engineering free · CS50x, Harvard — how software is actually built free · Apprenticeship.gov — electrical, HVAC, plumbing and fitting programs that pay while you learn paid to train · edX — operations management and process monitoring courses free to audit · MIT OpenCourseWare — circuits, systems and diagnostic method, free and ungated free · Coursera — people management and team leadership specialisations free to audit · Purdue OWL — the standard reference for professional writing free · edX — performance measurement and evaluation free to audit

All 35 skills ranked by how many jobs they open →

Where this experience transfers — occupations you could move toward

Computed from U.S. Dept. of Labor O*NET skill and knowledge profiles: high overlap with what you already do, a materially higher resistance score, no large jump in required training, and no licence you would have to start a new pipeline to get. Targets that pay meaningfully less, that are themselves COOKED, or whose own headcount is falling are excluded — a move into a shrinking trade is not an escape.

Database Architects EXPOSED · 38/100 · you already have ~86% of the skill profile

Skills to close: Systems Analysis

Software Developers EXPOSED · 41/100 · you already have ~81% of the skill profile

Skills to close: Technology Design, Installation, Operations Monitoring, Troubleshooting

Computer Occupations, All Other EXPOSED · 34/100 · you already have ~73% of the skill profile

Skills to close: Management of Personnel Resources, Writing, Monitoring

What would move this back up — beyond any one person

The moves above are yours to make. This is the other half: what would have to change in the world for the occupation itself to score higher. None of it is in any one person's gift, but it is where the floor actually comes from. Scores here are not a one-way ratchet. Only two of the five dimensions — task resistance and embodiment — track what machines can do. The other three track law, what buyers will pay for, and who is answerable, and those move in both directions, often in response to the same pressure AI creates. If every lever below landed, this occupation would score around 39/100 — EXPOSED.

4 specific changes that would raise this score
  • already happening task resistance +4

    Task-mix shift, already visible: the routine tier (markup, CMS theming, responsive layout, form wiring) is generated in one pass, leaving the tier models fail at — reproducing a Core Web Vitals regression in a live system, screen-reader testing against real assistive tech, CSP and supply-chain hardening after a third-party script compromise, and migration of legacy production data without downtime. This raises the score only for developers who own production systems; site-builders have no second tier to fall back into.

  • plausible liability shield +5

    Accessibility law hardening: if DOJ's ADA Title II web rule (28 CFR Part 35, WCAG 2.1 AA, compliance deadlines 2026-27) is extended to Title III private commercial sites, or the European Accessibility Act's 2025 enforcement produces named-signatory conformance statements, a human would attest to VPAT/accessibility conformance and carry personal or firm liability for the claim. Similarly PCI-DSS 4.0's client-side script inventory requirements (6.4.3/11.6.1) already name an accountable party for payment-page JS.

  • plausible judgment accountability +4

    If on-call and incident ownership becomes the formal core of the role — named incident commander in SOC 2 / ISO 27001 audits, breach-notification decision authority under state privacy laws — the role owns consequential calls under ambiguity rather than producing artifacts. Insurer cyber-policy underwriting increasingly demands a named technical owner for web-facing assets.

  • unlikely trust premium +2

    Narrow route only: buyers paying for provenance of code they can be sued over — a human-authored, human-reviewed attestation for regulated sectors (health portals, government procurement, financial disclosure pages). No plausible route for the general small-business site market, where price is the only variable and AI-generated sites are already the default.

The limit. Even with every lever, this stays a low-to-mid score. There is no licensure for web development anywhere in the US, and none is being proposed by any state board or professional body, so liability_shield cannot approach the levels seen in engineering or medicine — the accountability attaches to the firm, not the individual. Embodiment has no route at all. The realistic ceiling is roughly 40-45, and it applies to developers who own production systems, not to the modal agency site-builder described here.

These are conditions, not forecasts — what would have to happen, not what will. Specific rules, cases and bills are named so you can go and check whether they exist and where they stand; verify before relying on any of them. Nothing here is legal or financial advice.

Where this work is, and what it pays there

BLS metro figures for 163 areas. The verdict above does not change by city — the rubric judges what the work involves, not where it happens — but pay and headcount do, and the national median hides a very wide range.

Most of these jobs

New York-Newark-Jersey City, NY-NJ 4,200 $104,820 +13%
Washington-Arlington-Alexandria, DC-VA-MD-WV 3,840 $134,350 +45%
Seattle-Tacoma-Bellevue, WA 3,470 $130,440 +41%
Chicago-Naperville-Elgin, IL-IN 2,790 $96,870 +5%
Los Angeles-Long Beach-Anaheim, CA 2,310 $107,290 +16%
Dallas-Fort Worth-Arlington, TX 1,660 $96,740 +4%
Boston-Cambridge-Newton, MA-NH 1,450 $108,520 +17%
San Francisco-Oakland-Fremont, CA 1,420 $152,450 +65%

Best paid

San Jose-Sunnyvale-Santa Clara, CA 1,130 $166,670 +80%
San Francisco-Oakland-Fremont, CA 1,420 $152,450 +65%
Washington-Arlington-Alexandria, DC-VA-MD-WV 3,840 $134,350 +45%

Percentages are against this occupation's national median of $92,650. Counts are jobs in that metro, not vacancies. Metros where the BLS suppressed the cell are absent rather than shown as zero.

Who is actually doing this — nobody, on the record

We have no reported case of a named organisation automating this occupation. Not one deployment, not one announcement.

That is worth saying out loud next to a score of 24. The verdict above is about what the work exposes — what current AI could do to these tasks. It is not a claim that anyone has done it. For this occupation those two things have come apart completely: the capability argument is on this page, and the evidence column is empty.

Read that as a gap in the reporting we can see, not proof of absence — the dispatch runs on English-language feeds and misses plenty. If you know of a case, tell us, or add a field report from inside the job.

Quick take — do you do this job?

Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.

Self-reported and unverified — a sentiment signal, not a survey. One response per person per occupation; you can change your answer.

Field reports — what people say has changed

No field reports yet. A written account takes a paragraph rather than a tap, goes to an editor before it appears, and is the one thing on this page the rubric cannot produce on its own.

File a field report

Concrete beats general: a tool that arrived, a task that moved, a headcount decision you watched happen. Don't include anything that identifies you or your employer if that would put you at risk.

Watch this verdict
Kept current

Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.