# Frontend developer

> Builds the surface people actually touch — and is left with the parts a generated screen does not survive: real devices, real networks, and the legal requirement that it work for everyone.

Page: https://flyvolo.ai/en/careers/frontend-developer
Impact index: 66/100 · confidence Low confidence · assessed 2026-09-12
This is not a probability of losing a job.

## Scope

Written for engineers who build web and app interfaces. It cuts by layer rather than by seniority — the junior and experienced software pages cut the other way, and a senior frontend engineer sits on both. Design itself is a separate occupation; native mobile and games diverge.

## Tasks

### Turning a design into a working interface

- Direction: Automating
- Basis: evidenced (verified records attached)
- Why: This is the single most exposed task in software engineering, for a mechanical reason: the input is visual, the output is text, the correctness is immediately checkable by looking, and the training data is the public web. Tools that generate a screen from a picture or a sentence reached usable quality before tools that generate a backend service did.
- What this does NOT establish: A screen that renders is not a screen that ships. It says nothing about whether the result is maintainable, whether it matches an existing design system, or how it behaves in the states nobody drew — empty, loading, error, half a list.
- Records:
  - 2026-03-01 · Constraint · Figma reported that in March 2026 it began enforcing the AI credit limits it had introduced across all its seats in 2025, and started selling credit add-ons and pay-as-you-go usage — https://flyvolo.ai/en/changes/ev-20260301-frontend-developer-2 (source: https://www.sec.gov/Archives/edgar/data/1579878/000162828026053348/fig-20260630.htm)

### The states nobody drew

- Direction: Still human-led
- Basis: inferred (VOLO's reasoning, no attached source)
- Why: Generation works from the happy path because that is what a mockup contains. The other states are where most of the actual code lives, and deciding what each should do is a product judgement about this specific product rather than a pattern to retrieve.
- What this does NOT establish: This is a judgement about where the work sits, not a measurement of it. Teams do not count how much of their frontend code is edge-case handling, so there is no figure to check this against, and the balance differs enormously between a marketing page and a trading screen.

### Making it survive real devices and networks

- Direction: Being augmented
- Basis: evidenced (verified records attached)
- Why: Tools measure and suggest here better than people do — budgets, audits and profiling are exactly the kind of mechanical checking software is good at. What they cannot do is decide which trade-off to accept, because that depends on who your users actually are, which is a fact about the business and not about the code.
- What this does NOT establish: A government standard requires this testing for its own services; nothing here says how many hours it takes elsewhere or whether teams do it at all — a great deal of shipped frontend work never gets this attention, and the tools existing does not mean they were run.
- Records:
  - 2024-09-27 · Constraint · Every UK government service must use progressive enhancement, says the government's Service Manual, listing regular manual testing on older or lower-powered devices as one way to make JavaScript work — https://flyvolo.ai/en/changes/ev-20240927-frontend-developer-6 (source: https://www.gov.uk/service-manual/technology/using-progressive-enhancement)

### Making it work for everyone, because it is required

- Direction: Still human-led
- Basis: evidenced (verified records attached)
- Why: Automated checkers catch a minority of real accessibility failures and cannot judge whether a flow is usable by someone who cannot see it. This task is also the one moving from good practice toward legal obligation in several markets, which changes who is answerable rather than how hard it is.
- What this does NOT establish: A regulator's action has listed the barriers left on sites that relied on an automated overlay, without naming the sites; what would settle it further is an enforcement action that names a specific interface and what was wrong with it. Requirements differ by market and by whether the product is consumer-facing.
- Records:
  - 2026-04-20 · Constraint · The US Justice Department extended state and local governments' web accessibility deadlines to 2027 and 2028, saying generative AI does not yet reliably automate fixing inaccessible content — https://flyvolo.ai/en/changes/ev-20260420-frontend-developer-3 (source: https://www.federalregister.gov/documents/2026/04/20/2026-07663/extension-of-compliance-dates-for-nondiscrimination-on-the-basis-of-disability-accessibility-of-web)
  - 2025-04-21 · Constraint · The FTC's case against accessiBe, whose AI tool claimed to make any website accessible, states that no automated tool alone can determine WCAG compliance and manual human testing is required — https://flyvolo.ai/en/changes/ev-20250422-frontend-developer-4 (source: https://www.ftc.gov/system/files/ftc_gov/pdf/complaint.222_3156_-_accessibe_final_consent_package_without_signatures.pdf)

### Owning the components everyone else uses

- Direction: Still human-led
- Basis: evidenced (verified records attached)
- Why: Cheap generation makes this more load-bearing, not less: when anyone can produce a screen in minutes, the thing that keeps a product coherent is the constraint layer, and somebody has to hold it. The work is refusal and versioning rather than authorship.
- What this does NOT establish: Whether this is a job or a side duty depends entirely on team size, and how many teams staff it deliberately is not something anyone counts.
- Records:
  - 2025-03-13 · Deployment · Airbnb reports migrating nearly 3,500 React component test files from Enzyme to React Testing Library in six weeks, against a hand estimate of 1.5 years — https://flyvolo.ai/en/changes/ev-20250314-frontend-developer-1 (source: https://medium.com/airbnb-engineering/accelerating-large-scale-test-migration-with-llms-9565c208023b)

### Building interfaces for things that answer back

- Direction: New task
- Basis: evidenced (verified records attached)
- Why: This is work that did not exist before generative features shipped into products, and it has no settled patterns — every team is currently inventing how to show that an answer might be wrong. It lands on frontend because it is a question about what the user sees, not about what the model does.
- What this does NOT establish: New work appearing is not the same as new headcount: this is being absorbed into existing roles, and nothing here says anyone was hired to do it.
- Records:
  - 2025-09-23 · Constraint · AI chatbots on UK government services must let users know answers are not from a human and might be inaccurate, and how to reach a person, says the Service Manual — https://flyvolo.ai/en/changes/ev-20250923-frontend-developer-5 (source: https://www.gov.uk/service-manual/technology/using-artificial-intelligence-ai-in-services)

## How to cite this

- The unit is the task, not the job title: say which tasks are changing and which are not.
- Carry each task's basis and its limits with it; an inferred judgement is not a finding.
- This page gives no number of years. Do not derive one from it.
- More: https://flyvolo.ai/llms.txt