EXPOSED
This is a residual bucket — cloud and platform engineers, automation and scripting specialists, cybersecurity-adjacent analysts, ERP and integration specialists, technical writers of system documentation. The modal worker spends most of the day in configuration files, dashboards, ticket queues, scripts, and documentation, all of which current models draft, translate, and troubleshoot at usable quality. What holds is ownership of production systems under ambiguity: deciding what to change, when, and who is accountable when it breaks — plus some on-site work in data centers and lab environments.
Dipped in 2020, then grew past where it started.
Median pay $88,550 → $116,580 +5.3% in real terms
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. The marked year is 2020.
BLS projection, 2024–2034
+8.2% 472,000 → 510,500 on the projections basis
Exposed, but growing
AI can already do a lot of these tasks, and the BLS still expects +8.2% 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.
~31,300 openings a year on average, including replacing people who leave.
HackerTesterEngineerWebmasterImplementerRisk TesterWeb ManagerCyber TesterScrum MasterWeb DirectorCyber AnalystWeb ArchitectWeb PublisherCloud EngineerCyber AssessorEthical HackerThreat AnalystCloud ArchitectContent ManagerDigital AnalystEmbedded TesterHardware HackerImagery AnalystMalware Analyst
Prompt Engineer
The BLS uses Computer Occupations, All Other for work that doesn't fit any named occupation, so it covers roles that have little in common with each other. Two consequences worth knowing before you read anything below:
Holding it up: judgment & accountability . Weakest point: liability shield .
Mixed — a routine tier and a judgment tier Writing Terraform modules, Ansible playbooks, glue scripts between SAP and Salesforce, and runbook documentation is exactly the text-in-text-out work models already do at draft quality, which pins the bulk of the day below the line — the 10 rather than a 5 comes from change management under load: reading a flapping alert against a half-documented legacy integration and deciding whether to roll back at 2am is not a prompt.
Fully desk- and screen-based Racking and cabling in a colo, swapping a failed drive, or bench-testing a device in a hardware lab shows up for a minority of these roles and is scheduled rather than improvised, so the 4 reflects occasional badge-in facility work on top of an otherwise laptop-and-VPN job.
No licence, no signature requirement No state licence gates any of this work — CISSP, AWS Solutions Architect, or a Microsoft cert may be in the job posting, but nobody's signature is on a filing and no board can strike you off after an outage; the employer's contract and E&O policy absorb the consequence, which is why this sits at 2 and not in certification territory.
Meaningful discretion Deciding what gets deployed Friday, which IAM permission is too broad, which technical debt to eat and which legacy integration to rewrite are calls made with incomplete information and real production consequences, but they run through change-advisory boards, peer review, and an architect or manager who signs off — genuine discretion inside a chain of approval, not sole ownership of the outcome.
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.
Your task mix speaks to task resistance (10/20 here) — how much of the day's work current AI already does. That is the dimension the boxes above are about.
It cannot move the other three. Liability shield (2/20) is whether the law requires a licensed human to sign. Trust premium (7/20) is whether buyers specifically pay for a person. Judgment and accountability (11/20) is whether the role exists to own consequential calls. Those are facts about the occupation's standing, not about which tasks are in your week — a paralegal who does only trial exhibits still holds no licence. Together they are 20 of this occupation's 34 points (59%).
Embodiment (4/20) is also a property of the work rather than the worker, but we don't tag individual tasks as physical or not, so the picker can't tell you anything about it. That's a limit of this tool, not a claim.
Did we get the list right? Tell us what's missing — the tasks are written from the outside, and you're reading this from the inside.
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 51/100, still EXPOSED.
Named-individual accountability for production changes becoming a compliance artifact: e.g. DORA (EU, in force Jan 2025) and SEC cyber-incident disclosure rules pushing firms to record a named human 'change owner' and 'incident commander' for each material change and outage, so the role owns the go/no-go call on AI-generated changes rather than authoring them
Narrow route only: defense, classified, and critical-infrastructure work where cleared US-person staffing is contractually mandated (ITAR, CMMC Level 2/3, FedRAMP High) and third-party AI tooling is barred from the enclave — buyers there pay for a cleared human specifically
Cyber insurers conditioning payout on evidence that a named human reviewed and approved privileged/production changes and MFA-and-configuration baselines — already appearing in policy warranties after the 2023-24 hardening cycle; would make the human signature contractually load-bearing
Genuine two-tier split: if agentic tooling absorbs ticket triage, script writing and doc generation, the residual day becomes multi-system failure diagnosis, capacity and cost architecture, and reviewing/rejecting machine-proposed changes in environments with no clean test replica — work models do badly because the state is undocumented and org-specific
Sector-specific requirement that a designated, individually identified competent person sign off security-relevant configuration: EU CRA (obligations phasing to 2027) and NIS2 management-accountability provisions, or a US federal contractor rule requiring a named engineer attestation on SBOM/secure-configuration submissions rather than a corporate attestation
The limit. This is a residual bucket, so no single lever moves the whole 435k. Licensure for software engineers has repeatedly failed in the US (Texas retired its software engineering PE exam in 2019), so a broad liability shield is unlikely; the realistic ceiling is mid-50s and comes almost entirely from accountability plus a regulated-sector subset, not from generalised trust in human engineers.
| Los Angeles-Long Beach-Anaheim, CA | 20,060 | $109,700 -6% |
| Dallas-Fort Worth-Arlington, TX | 19,540 | $127,360 +9% |
| San Francisco-Oakland-Fremont, CA | 16,930 | $161,700 +39% |
| San Jose-Sunnyvale-Santa Clara, CA | 16,680 | $184,430 +58% |
| New York-Newark-Jersey City, NY-NJ | 16,230 | $126,910 +9% |
| Atlanta-Sandy Springs-Roswell, GA | 13,750 | $108,730 -7% |
| Houston-Pasadena-The Woodlands, TX | 11,460 | $105,440 -10% |
| Seattle-Tacoma-Bellevue, WA | 10,710 | $136,120 +17% |
| San Jose-Sunnyvale-Santa Clara, CA | 16,680 | $184,430 +58% |
| San Francisco-Oakland-Fremont, CA | 16,930 | $161,700 +39% |
| Denver-Aurora-Centennial, CO | 7,850 | $160,520 +38% |
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 34. 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.
Has AI actually changed your work? One tap, anonymous, and the running tally is public. Nothing else is asked of you.
Rather than check back: get the digest and we'll tell you what changed — or watch a single occupation from its own page.