Loading page content
Preparing content…
SOC 27-3042 · O*NET 27-3042.00
Technical writers research product specifications and communicate with technical staff, turning complex information into manuals, operating guides, online help, and illustrations that users can understand and execute, and maintaining terminology, versions, and revision records. The focus is whether the reader can complete the task under current conditions rather than elegant wording; writing steps does not mean having authority to approve product safety, medical, or engineering conclusions. The US code is 27-3042.00, and the 2025 wages and 2025–2035 projection are recorded per the precise occupation; China income and comparable recruitment samples remain a visible gap. You can first use this page's fictional notebook v2 specification to write a one-page quick start, failure branches, and a revision record and to do a paper walkthrough; the work does not pretend to be real product testing and does not disclose customer logs or unauthorized images and text.
Model impact index for three equally weighted representative core tasks; not a measured share of working time or a job-loss probability. floor((50+75+75)/3+0.5)=67; fermatmind-ai-impact-v1; 2026-10-18.
See whether the person can distinguish reader prerequisites, entry points, successful outcomes, and failure handling, and raises specific questions when the specification does not say. A smooth paper document does not mean real software usability has been verified.First compare the two internal task lines of user instructions and maintenance and revision, then compare editors, proofreaders, writers and authors, and document management specialists. Shared writing ability does not mean the posts can be directly exchanged.
This is career exploration and does not substitute for technical domain knowledge, employer requirements, engineering approval, or real usability testing.
Willingness to question users about goals, re-check product versions, turn vague descriptions into executable steps, and handle technical review feedback patiently is worth experiencing in technical writing first. Writing well is only one of the conditions.The more suitable behavioral signals are willingness to ask experts about conditions, check terminology, record changes, and proactively point out unknowns. When the document is revised, the person discusses around the reader's task rather than only defending their own sentences.
Technical writers research product specifications and communicate with technical staff, turning complex information into manuals, operating guides, online help, and illustrations that users can understand and execute, and maintaining terminology, versions, and revision records. The focus is whether the reader can complete the task under current conditions rather than elegant wording; writing steps does not mean having authority to approve product safety, medical, or engineering conclusions. The US code is 27-3042.00, and the 2025 wages and 2025–2035 projection are recorded per the precise occupation; China income and comparable recruitment samples remain a visible gap. You can first use this page's fictional notebook v2 specification to write a one-page quick start, failure branches, and a revision record and to do a paper walkthrough; the work does not pretend to be real product testing and does not disclose customer logs or unauthorized images and text.
Career direction comparison
First distinguish the two task lines of user instructions and maintenance and revision from independent occupations. Editors, proofreaders, writers and authors, and document management specialists may share text and material ability, but deliverable goals and responsibilities differ. When choosing, look at the output you can sustain long term and the actual gaps, and do not assume one guide lets you swap between all of them.
| Career direction | Core work and outputs | Key differences | Which choice suits better |
|---|---|---|---|
| User instructions | Arrange prerequisites, steps, successful outcomes, and failure handling for the target reader. | This is an internal technical writing task line requiring the current product version basis; it is not software development or safety approval. | Willing to break complex conditions into clear steps and clarify with technical experts. |
| Maintenance and revision |
AI impact on the occupation
First separate draft capability, actual adoption, document outcomes, and employment outcomes. O*NET defines tasks, BLS provides the US projection explanation, and NIST defines generic generation risks; this page's documentation check steps are editorial advice, with no single-occupation accuracy or time-saving measurement. AI Impact Score 67% is an editorial estimate. Model impact index for three equally weighted representative core tasks; not a measured share of working time or a job-loss probability. floor((50+75+75)/3+0.5)=67; fermatmind-ai-impact-v1; 2026-10-18.The three tasks, ratings and sources are listed in the FermatMind method source record on this page.
FermatMind AI Impact Score estimation method v1
FermatMind / 费马测试 · 2026-10-18
Chinese mainland salary reference
China salary definitions
Task scope and gaps
Insufficient evidence: a precise nationwide China technical writer wage distribution was not obtained. National Bureau of Statistics industry or broad category average wages cannot substitute for this occupation or serve as a personal take-home monthly salary.
Insufficient evidence: a precise nationwide China technical writer wage distribution was not obtainedNo usable precise nationwide distribution is currently available, so no estimated wage is entered.
No reviewable same-occupation employer sample has been formed, so recruitment volume, education and pay differences, or hiring probability cannot be computed from it.
US salary reference
The US May 2025 technical writer annual median wage is 90,390 US dollars and the hourly median is 43.46 US dollars. Both are official published definitions; they are not personal take-home pay or China pay.
| Wage position | Annual wage and hourly reference | Definition boundaries | Explanation |
|---|---|---|---|
| Lower wage position | 57,440 US dollars per year (27.62 US dollars per hour) | Official annual value; not annualized independently | The 10th percentile of 57,440 US dollars per year (27.62 US dollars per hour) describes a statistical low-end position, not a unified new-hire salary grade or hiring guarantee. |
| Nationwide annual median wage | 90,390 US dollars per year (43.46 US dollars per hour) | Not the same as take-home or a fixed monthly salary |
Career fit guide
Observe information research, reader-facing expression, version re-checking, technical questioning, and willingness to revise continuously at the same time. Personality tests at most provide discussion leads; this page uses specific behavior and paper works to verify the next exploration step.See whether the person can explain from the current specification, keep unknowns, and be willing to revise again. Introversion, extraversion, personality, or one test score do not substitute for actual documentation ability and target post knowledge.
Worth further exploration： Observe whether the person can rearrange instructions for the same user goal and point out the basis for each revision; do not judge occupational ability from one instance of fluent writing.
Check real conditions first： Check fixed employment and project writing separately for continuity, payment, and costs. Liking to write instructions does not mean you can bear gaps between projects, and budgets do not rely on an imagined next contract.
Try a small experience first： Use the v2 specification to write a one-page quick start and failure branches, re-check the step matrix, and then revise the v1 old draft. Paper only, and no real user testing is claimed.
Work pressure, risks, and career boundaries
The core risks are documentation following an old version or wrong product behavior, unclear image, text, and log use, confused technical approval, and uncontrolled maintenance scope. First check audience versions, expert responsibility, and contracts, then see whether the individual can sustain update communication.
These are editorial notes compiled from tasks, occupational background, general copyright and personal information rules, not accident rates, legal rulings, or employer scores. Specific product regulations and organizational review are checked separately.
Typical scenarios：When using others' images or text or works made in employment, only the download source is recorded, without checking licensing, contracts, and quotation scope.
Main positions：Writers producing instructions, illustrations, and public works
Possible consequences：May exceed permitted use and cause copyright or cooperation disputes.
Entry, certificates, and career development
First produce user instructions with a current basis, then practice maintenance revision and technical questioning; actual product work requires domain knowledge, real verification, and organizational review. The following stages are editorial ability organization, do not guarantee promotion by years, and do not substitute for real employer requirements.
Career outlook and related occupation transitions
US 2025–2035 technical writer growth is 0.8%, with OOH rounding to 1%; the 3200 annual average openings are mostly replacements and cannot be treated as annual net additions.Moving to editing, proofreading, creation, or document management requires proving the target output; technical writing experience can provide a material organizing foundation but does not automatically grant qualification or hiring.
United States · US SOC27-3042 precise occupation
0.8% (OOH states the integer 1%)
Overall employment change
BLS says digital product support expansion brings demand, and AI raising productivity may slow growth; the two coexist and are not a measured China pay cut or net AI headcount reduction.
Cannot show：A US total projection only, not extrapolated to enterprises, China, or individuals, and it cannot be split into AI causal headcounts.
Research the current specification and reader needs, write operating guides, help, and illustrations, and maintain terminology, versions, and revision records; technical facts and release responsibility are confirmed per the organization.
Technical writing centers on complex product instructions; editors plan, review, and revise content, and proofreaders mark errors against manuscripts. Shared text does not mean the same responsibility.
Choose a real task from this page for a small practice project or a conversation with someone in the field. Use the planning guides to set a date, an output and feedback questions before deciding whether to continue.
First check the audience and deliverable type, product domain, technical reviewers, version update cadence, image and text rights, and contracts. Document authorship, engineering confirmation, and release approval are different responsibilities.If someone only wants to create free-form text and is unwilling to re-check specifications, handle repeated updates, or accept multi-party review, compare the creation and editing directions first. Job conditions and support cannot be inferred from personality type.
Use the notebook v2 complete fictional specification below to write a quick start and two failure branches, mark revisions against v1, and then do a paper walkthrough per the rules. Do not download software, register accounts, or work with real data.Preserve four works: a one-page quick start, a step source matrix, a version change note, and a paper walkthrough. The next step is checking real job domain knowledge and documentation requirements, and honestly explaining the difference between the exercise and real project experience.
The following two lines are editorially summarized from technical writing tasks and are not a unified enterprise process. Each step still needs the current specification and the corresponding technical review.
Current-version instructions for a specific task and questions pending confirmation
Traceable version differences and updated help
27-3042.00 technical writers write technical materials such as manuals, operating and maintenance instructions, and appendices, and may assist with layout. Content editing, proofreading, and enterprise document system management each have independent occupational scopes. Both must check the technical basis and reader needs. New instructions first establish the task and structure; maintenance first locates change impact and old-version errors. The table below is an editorial comparison and does not claim all enterprises use the same process.
| Dimension | Main directions | Comparison directions |
|---|---|---|
| Purpose | Help the target reader complete a current product task. | Keep existing documentation accurate with versions and feedback. |
| Input | Current specification, reader conditions, product samples, and expert explanations. | Change records, old documents, feedback, and the current specification. |
| Process | Extract prerequisites and arrange steps, outcomes, and failure handling. | Locate affected passages, update entry terminology, and check related links. |
| Output | A quick start, user guide, or online help draft. | A revised version, change note, and pending list. |
| Responsibility | The author organizes the expression, and technical experts confirm domain facts. | The author records changes, and the responsible person confirms the version and release. |
| Cooperation | Ask product, engineering, and user support about conditions and goals. | Coordinate change scope and timing with product, engineering, and support. |
| Boundaries | Writing a guide does not equal completing real usability testing. | A revision record does not equal product safety or release approval. |
Determine the task, reader prerequisites, and current product version, research sample specifications, and clarify conflicts with technical staff.
Arrange instructions by order, clarity, and concision, keeping entry names, objects, and terminology consistent.
Write operation and maintenance materials or online help, and choose illustrations that support understanding and check their source and use.
Edit and standardize others' materials, and review facts, limitations, and reader questions with domain experts.
Monitor technical changes, keep revision records, and confirm that affected steps and links are updated.
Assist with layout, publishing, and distribution per the actual division; release authority and technical approval are checked separately.
The following are editorial self-check questions for this page's paper exercise; real products also need technical review and actual verification.
Organize materials by order, clarity, concision, style, and terminology standards, interview technical staff, research drawings and samples, write maintenance help, and track technical changes.
Inputs include the current product specification, version changes, reader needs, sample illustrations, and technical staff feedback. When materials conflict, return to the technical lead for clarification, and do not treat fluent generated text as fact.
The entry point, prerequisites, and result of the same step change with the product version; old documents cannot just have the date changed. A fictional paper walkthrough can find internal inconsistencies but does not prove competence in real software or equipment operation.
Write only the given failure symptoms and handling; retry counts, recovery mechanisms, and permissions not stated go to technical staff as questions, and are not invented.
| Track changes and feedback and correct document terminology, entry points, and related help. |
| Still technical writing, with a focus on change impact and version consistency; it is not enterprise document system management. |
| Willing to do repeated checking, keep revision basis, and coordinate update responsibility. |
| Editors | Editors plan, coordinate, and revise text and may review publication proposals; technical writers focus on product technical information and executable instructions. | Focuses on text planning, review, and publication choices, and requires separate proof of content judgment and editorial cooperation. | More willing to study structure, style, and publication needs and to accept author communication. |
|---|
| Proofreaders and copy markers | Proofreaders find grammar, typographic, and layout errors in manuscripts and proofs and mark corrections; content rewriting and technical fact confirmation are not the same responsibility. | Focuses on marking corrections against manuscripts, and rewriting and technical confirmation must go to the corresponding responsible persons. | Likes checking words, numbers, and layout item by item and verifying that corrections are implemented. |
|---|
| Writers and authors | Writers and authors create text such as stories, scripts, and advertisements; free creation or marketing output differs from instructions that follow technical specifications. | Original materials such as scripts, stories, and advertisements are the main output, and technical instruction samples cannot represent all creative ability. | Willing to research the topic, expression, and client and media requirements and to prove it with target works. |
|---|
| Document management specialists | Document management specialists implement enterprise document systems and management procedures; writing help material is not the same as configuring permissions, storing, or destroying systems. | Focuses on enterprise systems, access, and record processes, and needs system and management ability added. | More attentive to search, classification, permissions, and record procedures, and checks target requirements independently. |
|---|
If you like turning specifications into actionable steps, start with user instructions; if you prefer tracking changes and consistency, try maintenance and revision first. Moving to editing, proofreading, or creation requires target works; moving to document management requires system and process ability. One paper experience only helps narrow the scope, and real post conditions still need checking.
After choosing a documentation task, see which step AI changes, and separate generatable text from technical verification, actual adoption, and employment outcomes.
Editorial ratings use the existing occupational responsibilities and AI task analysis on this page. They are not adoption rates, measured effects, job-disappearance or personal-unemployment probabilities, or official scores from Stanford or another external institution. No measured time weights or future robot deployment are assumed.
Editorial advice: on permitted fictional or approved material, let tools propose structure and question drafts, then check each item against the current specification. Generating more sentences does not prove readers finish faster and does not exempt technical review.
Insufficient evidence: occupation-level actual AI adoption rates and verifiable net time savings for technical writers were not obtained. Tool capability, one team's adoption, and national usage rates each need evidence.
BLS2025 through 2035 US SOC27-3042 technical writer overall projection: 2025 base period 46,400 persons, 2035 end period 46,700 persons, official net increase 400 persons, growth 0.8%; annual average openings 3,200 including turnover exits and replacements, not an annual net increase or current recruitment. A precise occupation overall multi-factor projection, not an independent AI causality measurement. Headcounts are converted from thousands to preserve the original table precision, and the official net change is not rewritten as 300 by subtracting rounded endpoints; it includes 1.0% self-employed and a population different from OEWS employee wages, and does not infer China demand, real document accuracy or time savings, contract income, or personal unemployment probability.
| Work direction | Task | Current changes | Human control points |
|---|---|---|---|
| Structure | Draft the instruction structure per audience goals | Editorial advice purpose | Check the specification item by item and do not add features that do not exist. |
| Terminology | Propose glossary and consistency candidates | Editorial advice purpose | Confirm with product naming and domain experts. |
| Steps | Draft prerequisites, steps, and outcomes | Needs human checking | Check entry points, order, versions, and failure conditions. |
| Revision | Compare the old draft with change material | Editorial advice purpose | Distinguish real changes from model rewrites and preserve sources. |
| Material | Summarize approved technical material | Data use must be checked | Real personal information and undisclosed specifications cannot be uploaded at will. |
| Verification | Walk through whether the document is executable | Does not substitute for real testing | Check consistency on paper, and verify real effects separately. |
| Career choice | Judge tool and post demand changes | Causal net impact not quantified | Keep the US projection separate from the real China contract. |
The following evidence each answers different questions: the occupational material defines tasks, the risk framework flags fictional and privacy issues, and the US outlook indicates demand direction. Without supporting material, fluent generated text is not written as verified accurate, nationally adopted, or as having caused pay cuts in China.
Can generate quick starts, terminology candidates, and revision difference drafts; entry points, prerequisites, failure symptoms, and product versions still need human checking. Models must not add product behavior that was not provided.
What AI mainly changesGenerate structure and step candidates, with the basis still returning to the specification.
Still the responsibility of peopleClarify reader prerequisites, check entry points and outcomes, and submit for technical review.
What AI mainly changesPropose difference candidates, and old-version mixing is still possible.
Still the responsibility of peopleLocate real changes passage by passage and maintain terminology, links, and records.
What AI mainly changesFluent text cannot replace domain correctness verification.
Still the responsibility of peopleExperts confirm product facts, and the release responsible person confirms the version.
First confirm data use and the receiving tool, and use fictional product input. Real customer logs, undisclosed designs, and personal information must be confirmed per the organization and applicable rules, and are not uploaded just because an account can access them.
Clarify the audience, product version, and this document's scope.
Confirm data source use and practice with fictional input.
Return every claim to the specification, and hand conflicts to the technical lead.
Check prerequisites, entry points, outcomes, and failure branches step by step and do not fill in unknowns.
Preserve correction sources, versions, and approval status, with release confirmed separately.
Instructions may teach readers to use entry points or recovery methods that do not exist.
Control method:Bind every step to the current specification and keep missing items as questions.
Logs, screenshots, or undisclosed specifications may exceed their original use.
Control method:Check tool receiving and data scope, and keep exercises entirely fictional.
Users cannot complete the task in the current interface using the old draft.
Control method:Mark versions and locate entry points and prerequisites per the change.
A model's confident tone does not equal technical safety or public permission.
Control method:Keep confirmation status for authors, experts, and release responsibility separately.
Generation speed does not equal national job losses or pay cuts in China.
Control method:Check the research object and outcome indicators, and keep gaps publicly visible.
Use the notebook v2 fictional specification to write a guide and failure branches and preserve step basis.
Keep real changes separate from model candidates and record technical review and maintenance scope.
Check the real product domain, deliverable maintenance, and contracts, and do not treat the US projection as a personal outcome.
Check image, text, and log use and show only your own approved contribution; mark fictional exercises honestly.
It cannot be published based on fluency. Check every step's entry point, prerequisites, and outcome and the failure handling against the specification, mark unconfirmed behavior, and this page's paper exercise only verifies internal consistency.
Source:O*NET technical writers 27-3042.00Original material ↗
Not the same. Author time savings, reader completion, error rates, and maintenance cost are different outcomes; this page has no measurement and does not treat generation time as user effect.
Source:NIST generative AI risk overview AI600-1Original material ↗
No. Product facts and domain safety conclusions need confirmation by the corresponding responsible persons, and authors check expression and versions; the task profile only describes duties and grants no approval.
Source:O*NET technical writers 27-3042.00Original material ↗
Version conditions cannot be skipped: models may follow old entry points or mix platform features. State the product version, reader prerequisites, and material date in the text, and keep unstated recovery mechanisms as questions.
Source:O*NET technical writers 27-3042.00Original material ↗
There is no separately attributed net headcount. The BLS 0.8% total growth and the AI direction explanation cannot be split into losses already incurred, let alone represent China or an individual.
Source:BLS technical writer occupation outlookOriginal material ↗
No. NIST is generic risk material; real images, text, logs, and personal information use requires separate regional regulation and organizational agreement, and a model does not decide permission.
Source:NIST generative AI risk overview AI600-1Original material ↗
Write a one-page instruction from the complete fictional specification, then check the source version and failure branches step by step and preserve unknowns; record your own changes and do not pretend it is real software or user testing.
Source:NIST generative AI risk overview AI600-1Original material ↗
Authors bear expression and revision records, domain experts confirm product facts, and the publisher confirms the public version. Model summaries cannot substitute for technical, safety, or medical approval.
Quote and sample gaps
| Task direction | Data status | Applicability and gaps |
|---|---|---|
| User instructions | No comparable quotes for this direction at present | Technical writing focuses on user guides and help; comparable China quotes for this direction are currently missing. |
| Maintenance revision | No comparable quotes for this direction at present | The maintenance direction requires checking update scope and rounds; developer engineer pay cannot substitute for writer pay. |
| Freelance | No comparable project receipts at present | Freelance work requires checking deliverables, maintenance, payment, and costs; US employee wages are not treated as China project prices. |
| Nationwide technical writers | No precise nationwide wage distribution at present | The nationwide distribution gap cannot be filled by a few mixed post notices or industry averages. |
Keep the gap when there is no current comparable quote. The post name, publication date, tasks, location, employment relationship, and quote range must all be checkable; historical or adjacent post pay cannot be moved into technical writing.
When comparing contracts, first list the work domain, deliverable length and type, technical review, revision rounds, maintenance period, pay unit, and payment terms. Look at fixed post wages and per-project quotes separately.
Freelance project receipts also include business costs and non-project time, so one quote cannot be multiplied by the months of the year. Self-employed business receipts, pre-tax wages, and disposable income are checked separately.
The time scope, update responsibility, and payment milestones of long-term maintenance need to be listed separately; completing one piece does not mean a full-year contract continues.
Insufficient evidence: income changes and causal magnitude for China technical writers after using AI were not obtained. The BLS US outlook says AI may slow employment growth, which is not a China pay-cut measurement; revision time savings also do not mean contract pay changes in the same proportion.
| The median means about half of the statistical population is on each side; it is not a mean, new-hire pay, or a personal take-home guarantee. |
| Higher wage position | 145,270 US dollars per year (69.84 US dollars per hour) | A high percentile is not a personal income guarantee | The 90th percentile of 145,270 US dollars per year (69.84 US dollars per hour) describes a high-end threshold and does not represent a person's seniority or sustainable project income. |
|---|
Pay · 10th percentile：57,440 US dollars per year (27.62 US dollars per hour)；Precise occupation 2025 P10, not a new-hire grade.
Pay · 25th percentile：70,880 US dollars per year (34.08 US dollars per hour)；Precise occupation 2025 P25, not a seniority level.
Pay · median：90,390 US dollars per year (43.46 US dollars per hour)；Precise occupation 2025 median, not a mean.
Pay · 75th percentile：115,510 US dollars per year (55.53 US dollars per hour)；Precise occupation 2025 P75, not a senior grade.
Pay · 90th percentile：145,270 US dollars per year (69.84 US dollars per hour)；Precise occupation 2025 P90, not a personal guarantee.
This page uses official published annual wage values directly, and hourly data is listed only as a reference without multiplying by fixed hours. Individual paid arrangements and contract payments are checked separately.
OEWS wages exclude the self-employed, and the employment projection and the wage survey have different coverage. Self-employed pay cannot be inferred from employee percentiles, let alone China freelance net income.
Percentiles are not seniority grades. Product domain, duties, region, and personal conditions may differ, and a wage position cannot be directly labeled junior, intermediate, or senior.
Check the employment relationship, currency, pre-tax or post-tax status, deliverables and maintenance, revision scope, and payment cycle. Fixed employee wages and total project quotes are compared separately once the definitions are clear.
| Industry | Annual median wage / statistical status | How to understand |
|---|---|---|
| Professional, scientific, and technical services | 94,520 US dollars per year | The 2025 annual median for technical writers in professional, scientific, and technical services is not the average wage of all workers in that industry. |
| Information | 86,400 US dollars per year | The 2025 annual median for technical writers in the information industry is not the pay of all software engineers. |
| Manufacturing | 82,860 US dollars per year | The 2025 annual median for technical writers in manufacturing; it is not a specific factory quote or China manufacturing pay. |
| Self-employed technical writers (noted separately) | Statistics exclude the self-employed | Self-employment is an employment status and is not covered by OEWS wages; this row only notes the gap and does not take part in industry wage rankings. |
Fixed employee and freelance arrangements differ, and working hours, maintenance period, revision rounds, and payment milestones must all be confirmed per the contract.
Writing only a first draft, being responsible for technical verification, and maintaining multiple versions are different deliverable scopes; list the scope before comparing, and do not estimate quotes from an occupation name.
Office location, collaboration time zone, and pre-deadline work arrangements affect what the individual can bear, and an industry median cannot decide that alone.
2025 base period； wage and employment coverage are checked separately.
The 2025–2035 original value is 0.8%, and the OOH integer is 1%.
Includes growth and replacements, not 3200 net additions each year.
Can CAI decide whether I am suited to technical writing?
Standards organization, written expression, and information research can serve as interest clues; CAI is a US occupation profile, not a type the individual must match. Observe whether the person is willing to re-check product facts and revise repeatedly for readers.
Do not use it this way:Similar code does not mean sufficient technical domain knowledge and does not substitute for real working conditions.
Explore my career interestsWhich habits should be turned into practical questions?
See whether the person completes agreed documents, states the basis when finding conflicts, accepts revisions, and keeps records. There is no evidence for setting a technical writer pass line at any dimension level, and specific behavior is more useful than labels.
Do not use it this way:Check the real product release cadence, collaboration time zones, and deadlines, then assess bearable arrangements. Office work does not mean every post can be remote, and an occupation name does not imply mandatory overtime at any time.
Explore my Big Five traitsDo introversion or extraversion decide writing ability?
No. Independent organizing and cross-disciplinary questioning are different situations, and liking social contact does not mean being good at checking technical facts; a quiet person can also ask clear questions and complete documents.
Do not use it this way:Do not infer all application results from four letters or exclude product domains.
Explore my personality preferencesDo I like written expression or technical instruction work?
Distinguish liking to create sentences from being willing to research unfamiliar products, handle details, and process repeated updates. Motivations can coexist; observe whether the person can still check the basis around reader goals when feedback is unfavorable.
Do not use it this way:Do not infer others' real psychology or your own talent from motivation labels.
Explore my motivation cluesHow does comprehension land in the work?
Against the v2 specification, extract prerequisites, entry points, and success and failure conditions and find a basis for each instruction. When something is missed, locate the source item and step; a single general score does not substitute for this task, and there is no unified hiring line.
Do not use it this way:When an error occurs, check the material and practice conditions first, and do not infer overall ability from one omission.
Explore my cognitive performanceHow do you communicate with experts while keeping boundaries?
First restate the current version and reader goals, then ask answerable technical questions; unclear expression does not mean asking experts to write the entire draft for you. When personal information or public permission is unconfirmed, keep it as a question, and understanding others does not mean unconditional agreement to disclose.
Do not use it this way:You may refuse to disclose unauthorized screenshots, customer logs, and fabricated approval records. Proactively clarifying the material scope is a professional boundary, and determination to enter the field cannot be used to demand disclosure of real data.
Explore my communication cluesBetter match:Willing to write prerequisites, steps, outcomes, and failure handling for readers.
Note:Fluent expression does not prove real users can complete the task, and actual verification is still needed.
Better match:Willing to track changes, return to the old draft, and check related links.
Note:Changing the date does not mean the old entry points and terminology are updated.
Better match:More attentive to structure, style, publication judgments, and author cooperation.
Note:Target editing works need to be added, and seniority is not inferred from the number of guides.
View related occupations:EditorsBetter match:Willing to mark corrections item by item against manuscripts and proofs.
Note:Proofreading accuracy does not equal rewriting or technical approval ability.
View related occupations:Proofreaders and copy markersBetter match:Likes original expression in stories, scripts, or marketing material.
Note:Target media works are needed, and technical guides do not cover all creative work.
View related occupations:Writers and authorsBetter match:More attentive to enterprise file systems and search and permission procedures.
Note:System management ability needs to be added, and writing a manual does not mean being able to manage or destroy systems.
View related occupations:Document management specialistsReading comprehension first lands on prerequisites, entry points, outcomes, and exceptions. In this page's notebook v2 exercise, only cards pending archiving can be archived, the title must be non-empty, and success returns to the list showing archived; do not fill in the save entry point or automatic recovery of v1 from memory. When something is wrong, point to which specification and which step rather than broadly judging that you are not smart.
When told to write it clearly, first ask where the reader got stuck, what prerequisite is missing, and how the terminology is confirmed. You can revise item by item against this page's v1 old draft and the v2 specification. Accepting feedback does not mean changing technical facts unconditionally; conflicts go to experts for confirmation while preserving the original basis.
When the work is not accepted, check the target product domain, format requirements, and your own explanation, and do not guess the employer's sole reason. Record one step that could be improved and feedback not yet obtained; one application does not prove overall ability or personal worth.
Not the same. First try researching specifications, writing drafts, handling failure branches, and maintaining old versions, and record which tasks you are willing to repeat and which drain you. Interest in finished reading can be a starting point but cannot replace this kind of actual work experience.
First observe independent preparation and specific questioning ability. Communicating domain facts is required, but no fixed personality is demanded; being able to state versions, reader goals, and questions clearly is more useful. Real organizational communication load still needs checking.
No. A guide must match the product's current behavior; unknowns must be asked of experts, and rhetoric cannot cover errors or uncertainty. Writing a work from this page's specification can initially verify basis and expression but cannot prove mastery of a real engineering domain.
List the time slots, time zones, and deadline conditions you can accommodate, and compare specific posts. Constraints are not personality flaws; when information is missing, check the update frequency first and do not promise schedules you cannot deliver.
Yes. The notebook v2 specification, old draft, and answers are all on this page, and a paper step-by-step walkthrough is enough. If you did not actually run the software or find real users, write that it is paper work honestly, and do not treat reference answers as your own test evidence.
Turn the test description into small questions and then observe this session's steps, source matrix, and modification records. Keep actual behavior and unknowns, and do not retest until you get an ideal label.
List the completed instructions and failure branches, questions still to clarify, and willingness to keep re-checking, and choose only one follow-up task that can get feedback. If you dislike maintenance details, you can learn about adjacent creation or editing; do not promise career success from one experience.
Risk reduction：Public images and text require checking licensing, employment and commission, and quotation conditions and keeping checkable sources and uses. Ownership of work created in employment is judged specifically by law and contract, and the author's involvement does not automatically imply public permission.
Basis：Copyright Law of the People's Republic of China (2020 amendment)
Typical scenarios：The model or old draft gives definite steps, but the current behavior and safety conditions are not yet confirmed by experts.
Main positions：Authors writing equipment, engineering, or medical product instructions
Possible consequences：Readers may act on wrong instructions, and drafts cannot substitute for domain approval.
Risk reduction：Uncertain product behavior stays in the draft as a question and goes to domain experts for confirmation. Writers do not sign off real equipment, engineering, or medical safety conclusions based on documentation experience.
Basis：O*NET technical writers 27-3042.00
Typical scenarios：Only the first-draft length is discussed, without listing update periods, revision rounds, review, and payment conditions.
Main positions：Those in fixed posts or delivering documentation per project
Possible consequences：Continuous maintenance exceeds expectations, distorting income and time arrangements.
Risk reduction：Check fixed wages, per-piece quotes, maintenance contracts, and self-employed receipts separately; state revision rounds, deadlines, and payment milestones clearly, and do not promise personal full-year income from a US median.
Basis：BLS technical writer occupation outlook
Typical scenarios：The old draft still says save while the v2 entry point is now archive, and it still keeps the old immediate-completion description.
Main positions：Writers maintaining multiple versions or platform help
Possible consequences：Readers cannot complete the task in the current interface or misjudge the result.
Risk reduction：Mark product versions in the file and check every entry point, prerequisite, and failure branch per the change; changing the date on an old draft does not replace a complete re-check.
Basis：O*NET technical writers 27-3042.00
Typical scenarios：Uploading real material to revise a draft without checking processing purpose, lawful basis, and the receiving tool.
Main positions：Those handling customer screenshots and internal specifications
Possible consequences：May exceed the original use and cause privacy or technical data leakage.
Risk reduction：Exercises use only fictional specifications. Real logs, screenshots, personal information, and undisclosed technical material require checking processing purposes and tool receiving scope; a model-generated summary does not prove permission.
Basis：NIST generative AI risk overview AI600-1 · Personal Information Protection Law of the People's Republic of China
Typical scenarios：Receiving expression and technical opinions at the same time, and writing unconfirmed behavior as approved to meet the deadline.
Main positions：Writers accepting multi-party review and tight deadlines
Possible consequences：Document facts, responsibility, and version status become confused, causing repeated rework.
Risk reduction：Distinguish technical fact corrections, expression preferences, and release responsibility. Record the original basis and questions, land revisions in specific steps, and do not treat multiple rounds of comments as personality evaluation.
Basis：BLS technical writer occupation outlook · O*NET technical writers 27-3042.00
| Entry direction | What you mainly do early on | Employer requirements and observation basis | How to prove it without experience | Differences from other directions |
|---|---|---|---|---|
| User quick start | Check audience version and prerequisites, and arrange steps and outcomes. | Editorial observation: each step has a current specification basis. | Use the notebook v2 to write this page's quick start and source matrix. | It belongs to the technical writing task line and does not represent a real recruitment post. |
| Online help maintenance | Compare changes with the old draft and fix terminology, entry points, and failure branches. | Editorial observation: able to locate the v1 error and affected passages. | Preserve the old draft, the v2 revision, and the reason for each change. | Maintenance ability does not equal configuring an enterprise document system. |
| Illustration and layout cooperation | Choose illustrations that aid understanding and check their source and use. | Editorial observation: illustrations match the object version in the text. | You can draw fictional interface frames with text instead of using real screenshots. | Illustration cooperation does not equal a visual design occupational qualification. |
| Technical domain supplement | Ask about unfamiliar product limitations and avoid fabricating behavior. | BLS notes that technical knowledge helps; specific posts are checked separately. | List the recovery and permission questions the specification does not give and keep them as questions. | Paper exercises do not prove engineering, medical, or other professional competence. |
| Documentation project cooperation | State deliverables, maintenance rounds, payment, and review responsibility. | Editorial advice, not employer requirements obtained this time. | Write a note on the documentation scope and draft confirmation status. | A project contract does not represent unified pay or a hiring guarantee. |
This page did not obtain a same-occupation independent employer sample; official occupational background cannot pretend to be a real recruitment test.
Show the notebook v2 user goals, the basis for each step, and failure branches, not just polished sentences.
Against the v1 old draft, explain the changes in entry points, prerequisites, success, and failure, and keep unstated questions.
State this page's fictional input, what you wrote, and the paper walkthrough; do not fabricate customer authorization, real users, or software testing.
Write the goal and prerequisites, four steps, successful outcome, and two failure handlings per S1 to S4, and mark the notebook v2 as a fictional paper draft.
Row by row: prerequisite to S1, select pending archive to S2, non-empty title to S3, archive success to S3, archived cannot be archived again to S4, and wrong title returns to editing to S3. List unknown recovery mechanisms separately.
Preserve the v1 old draft and list five corrections: save to archive, objects limited to pending archive, non-empty title, success shows archived, and failure branches, marking the current specification for each.
Walk through the normal, empty title, and already archived three paths and list expected results and sources; write your own omissions and revisions. No real software was run, so no user success rate is claimed.
| Scenario | What to check | How to observe | Cannot be inferred from this |
|---|---|---|---|
| Explain the task | Ask about reader goals, product version, length scope, and reviewers. | Show the prerequisites and the step source matrix. | Editorial advice, not a unified examination for all employers. |
| Receiving revisions | Sort comments into technical facts, expression, and unknowns. | Show the item-by-item revision basis from the v1 old draft to v2. | Finishing a revision does not mean the facts have been approved by experts. |
| Cooperative delivery | Check the maintenance period, rounds, payment, and public material scope. | List draft questions and confirmation responsibility. | An interview invitation does not equal a contract or public permission. |
| Explain the work | State the fictional input, your own contribution, and the paper walkthrough. | Point to specific steps and failure branches. | Paper does not equal real software and real user testing. |
| Qualification / preparation item | Who it mainly suits | Real use in job seeking | Whether it is a common entry threshold | Enrollment / application conditions | Investment considerations | Important limits |
|---|---|---|---|---|---|---|
| Typical education background | People who need systematic learning in written expression and technical communication. | The BLS US typical requirement is a bachelor's degree; technical domain knowledge helps. | It is not a unified China statutory threshold. | Check the target employer's education and specialty against your own background. | Learn about course content and feedback for actual gaps. | Education does not guarantee hiring or competence in a product domain. |
| On-the-job training | People who need to adapt to organizational style and products. | BLS describes short-term on-the-job training. | It is US occupational background, not a unified course certificate. | Follow enterprise task versions and review rules. | Check the training scope and actual feedback. | Do not substitute the Job Zone general experience years for specific training. |
| Voluntary professional certification | Job seekers whose target post actually recognizes it. | BLS notes that some association certifications are not mandatory. | It is not a unified US occupational requirement. | This page does not provide enrollment details for unverified certifications. | First check employer recognition, then current conditions and costs. | A certificate does not substitute for technical facts or work ability. |
| China image, text, and data use | Those handling works made in employment, illustrations, and logs. | Check law and contracts and keep the approved public scope. | Material compliance and occupational qualifications are different matters. | Check per employment and commission licensing and personal information use. | Keep checkable sources and confirmation for specific material. | Do not adjudicate individual cases from this page's general law. |
| Specific employer requirements | Those who already have a real target post. | Check the domain, language, tools, and experience against the current notice. | This page did not obtain a representative recruitment sample. | Preserve the original notice date, status, and tasks. | Invest in skills only for clearly identified needs. | Do not substitute adjacent developer post requirements for this occupation. |
| Original documentation works | Explorers lacking real project experience. | Can show steps, sources, revisions, and unknowns. | The work is not a qualification certificate or hiring proof. | Use only this page's fictional material and mark contributions honestly. | Finish the four paper works before deciding next steps. | Do not falsely claim customer projects, real testing, or public authorization. |
S1: the target reader uses the notebook desktop v2 for the first time; the paper simulation has the list open, with no account, network, or real software. The list has only card A, title weekly plan, status pending archive. The goal is to archive A and see archived. S2: v2 first select the pending archive card and click edit, and the details show a title input box and an archive button; other statuses are not handled.
S3: when archiving, if the title is non-empty the app returns to the list and A's status changes to archived. If the title is empty it stays in the details and prompts that the title cannot be empty; after typing weekly plan in the original input box and clicking archive again it succeeds. S4: an archived card cannot be archived again, and the archive entry point is unavailable in the list; if an archived card is selected, stop and go back to selecting a pending archive card. The specification does not give undo, recovery, batch, network error, or permissions.
The v1 old draft says: open the card, click save, all cards complete immediately; if it fails, click repeatedly. This draft is only for finding errors and is not the basis for v2. Write a one-page v2 quick start: title, version and readers, prerequisites, four steps, success, and two failure branches; mark each item with S1 to S4.
Then write five version revisions and do a normal, empty title, and archived paper walkthrough. Write questions for behaviors not provided, and do not add undo or automatic retry on your own; preserve your own contribution and the fictional statement.
Notebook desktop v2: archive the first card. Prerequisites: already in the list, A pending archive, title weekly plan (S1). 1 select A and confirm it is pending archive (S1/S2); 2 click edit to enter the details (S2); 3 confirm the title is non-empty, and if it needs changing type weekly plan in the title box (S2/S3); 4 click archive (S3). Success: return to the list, A archived (S3). Empty title: stay in the details and see that the title cannot be empty, then type weekly plan and click archive again (S3); already archived: the entry point is unavailable, so stop and go back to selecting a pending archive card (S4).
Revisions: save to archive S2/S3; all cards to pending archive S2/S4; add the non-empty prerequisite S3; complete to return to the list archived S3; repeated clicking to handling empty title and already archived separately S3/S4. The paper normal path succeeds; the empty title prompts first and succeeds after filling; the archived card has no second archive. Undo, recovery, and other errors remain as questions and are not written as tested.
Do only paper instructions based on S1 to S4, without running the product, registering accounts, or deleting real data; the v1 in the text is a deliberately given wrong draft. Readers may use different wording but must preserve prerequisites, entry points, outcomes, and failure conditions. The source matrix lists steps to specifications, the revision table lists old sentence to new sentence to basis, and the walkthrough record lists path to expectation to unknown; real product and user performance are not verified.
Read S1 to S4 and list the notebook v2, first-time readers, the archiving goal, and behaviors not provided. Output: a prerequisites and unknowns card.
List selecting pending archive, editing, non-empty title, archiving, success, and failure conditions. Output: a steps to S1 to S4 matrix.
Write the title, prerequisites, four steps, success, and two failure branches. Output: your own draft; do not copy the reference as your own testing.
List five corrections: entry point, object, non-empty, outcome, and failure handling. Output: old sentence to new sentence to current source.
Check the draft along the normal, empty title, and already archived three paths. Output: expected results and omissions; do not claim to have run the software.
Keep the guide, matrix, change note, and walkthrough reflection, and state the fictional source and your own contribution. Output: an explainable work package.
Check whether omissions and unknowns are honestly preserved and whether you are willing to revise again. Output: one follow-up feedback task and real conditions, without a personality pass score.
Typical duties：Write prerequisites, steps, and outcomes from one current product task and ask technical staff about gaps.
Promotion evidence：Has a step-by-step source matrix and complete success and failure conditions.
Qualification boundaries：Real domain requirements and data use are checked per post, and paper does not prove actual verification.
Next step：First get feedback on specific steps, then add domain knowledge.
Typical duties：Complete guides within the confirmed scope and check technical facts with engineering and support.
Promotion evidence：Preserve review comments, correction sources, and draft status.
Qualification boundaries：Document authors do not automatically have engineering safety or release approval authority.
Next step：Expand to more tasks based on real feedback.
Typical duties：Organize multi-task and terminology systems and check platform and version differences.
Promotion evidence：Able to explain the basis for structure choices and cross-passage consistency.
Qualification boundaries：Complex products need corresponding expert verification, and writing skill does not substitute for professional knowledge.
Next step：Add real information architecture and reader verification methods.
Typical duties：Manage more complex content and review cooperation, keeping reader goals and the public version aligned.
Promotion evidence：Show your own contribution, checkable technical review, and update records.
Qualification boundaries：Leading documentation work does not automatically grant engineering or management appointment.
Next step：Add works and cooperation evidence per the target duties.
Typical duties：Locate the old draft's entry point, terminology, and prerequisites from the current change.
Promotion evidence：Preserve the old and new comparison and the source for each item.
Qualification boundaries：Changing the date does not prove the content is current, and real data use is checked separately.
Next step：Complete the paper revision of the notebook v1 to v2.
Typical duties：Check change impact, related pages, and release scope.
Promotion evidence：Able to explain what has been changed, what is pending, and who confirms.
Qualification boundaries：Technical facts and publication dates are confirmed by the corresponding staff.
Next step：Add link and terminology consistency per feedback.
Typical duties：Distinguish product and platform versions and applicability conditions to avoid mixing.
Promotion evidence：Has checkable version scope, review, and update records.
Qualification boundaries：One version's test is not extrapolated to all platforms or real user effects.
Typical duties：Read the specification and list unfamiliar terminology and unprovided behavior as specific questions.
Promotion evidence：Distinguish original facts, author inferences, and items to confirm.
Qualification boundaries：Paper exercises do not prove engineering, medical, or other domain knowledge.
Next step：Learn the target product basics first, then choose real posts.
Typical duties：Ask technical staff around reader tasks and implement confirmed comments.
Promotion evidence：Has the original basis, modification records, and confirmation status.
Qualification boundaries：Model summaries cannot substitute for expert approval or legal permission.
Next step：Train clear questioning and terminology explanation.
Typical duties：Coordinate product, engineering, and support feedback and distinguish fact corrections from style suggestions.
Promotion evidence：Able to explain how disagreements were handled and limitations not yet confirmed.
Determine reader goals, the current specification, and questions.
Arrange prerequisites, steps, outcomes, and failure handling and mark the basis.
Fix entry points, terminology, and links per the change, and keep revision records.
Add product knowledge and complex documentation responsibility, and do not infer expert sign-off from years.
Generic risk framework · Cross-context generative AI
Fabrication and privacy risk flags
Risk method
NIST generic risks flag fictional and privacy issues and require human checking of specifications and use; the framework's existence does not mean national adoption rates or improved document outcomes.
Cannot show：The NIST risk framework does not measure technical writing adoption, time savings, or income.
China · China technical writers; no representative sample at present
Insufficient evidence: deduplicated China recruitment scale not obtained
Recruitment sample gap
Insufficient evidence: the current deduplicated China technical writer recruitment scale and a representative sample were not obtained. Developer posts that also write technical documentation are not counted as same-occupation demand, and occupational material is not an open-position notice.
Cannot show：Without a representative same-occupation sample, recruitment heat or demand rates are not computed.
These are ability adjacency relationships and do not mean automatic promotion or that no further training is needed.
Transferable：Information research, terminology, and text revision can help in understanding editing work.
Needs improvement：Add publication judgment, structure planning, and author coordination works.
View occupation detailsTransferable：Version re-checking and detail checking connect with proofreading.
Needs improvement：Add item-by-item correction marking against manuscripts and proofs and review ability.
View occupation detailsTransferable：Audience expression and research can serve as original text foundations.
Needs improvement：Add target media works such as scripts, stories, and advertisements.
View occupation detailsTransferable：Classification, versions, and material cooperation share a common foundation.
Needs improvement：Add enterprise system, permission search, and record procedure knowledge.
View occupation detailsTransferable：Explaining revision basis can help editorial communication.
Needs improvement：Add manuscript evaluation and publication choices, not only product step checking.
View occupation detailsTransferable：Proofreading focuses on detail and connects with version re-checking.
Needs improvement：Use manuscripts and proofs to demonstrate error finding, correction marking, and re-checking.
View occupation detailsTransferable：Writing research and original material can share some abilities.
Needs improvement：Need separate narrative or marketing works and target media feedback.
View occupation detailsTransferable：Material maintenance can help in understanding the document management background.
Needs improvement：Check enterprise systems and record permissions, and do not destroy real files on your own.
View occupation detailsUse this page's notebook v2 specification to write a quick start, source matrix, change note, and paper walkthrough; do not register accounts or pretend to have real product experience.
The US BLS typical requirement is a bachelor's degree, technical knowledge helps, and short-term on-the-job training is provided; association certification is not mandatory. China's specific post conditions are checked separately, with no unified certificate promise.
Insufficient evidence: this page has not obtained causal evidence of the net AI employment effect on technical writers or real documentation outcome evidence. BLS explains that AI productivity may slow US growth but does not provide a separate net AI headcount. Occupation adoption, accuracy, time savings, and China income causality data are insufficient, so no displacement rate is entered. BLS2025 through 2035 US SOC27-3042 technical writer overall projection: 2025 base period 46,400 persons, 2035 end period 46,700 persons, official net increase 400 persons, growth 0.8%; annual average openings 3,200 including turnover exits and replacements, not an annual net increase or current recruitment. A precise occupation overall multi-factor projection, not an independent AI causality measurement. Headcounts are converted from thousands to preserve the original table precision, and the official net change is not rewritten as 300 by subtracting rounded endpoints; it includes 1.0% self-employed and a population different from OEWS employee wages, and does not infer China demand, real document accuracy or time savings, contract income, or personal unemployment probability.
Insufficient evidence: this round did not obtain a nationally comparable wage distribution for the China technical writer occupation. No. US 2025 precise occupation wages are not China post quotes, after-tax income, or freelance receipts; the China national distribution remains a gap. China scientific research and technical services urban unit pre-tax annual average wages in 2025: 182,064 yuan per year for non-private and 83,560 yuan per year for private, with the two unit types counted separately in aggregate and not combined into a personal wage range. They cover legal-person units with 5 or more persons employed in urban areas, exclude self-employed individuals and freelancers, and include personal income tax, social insurance, and housing fund deductions; use only as background when the actual employer belongs to that industry, and manufacturing, information software, publishing, and other industries are not automatically included by technical documentation task names. This is not the median, starting pay, developer pay, per-project quote, or cost-deducted net income for the technical writer occupation, and it does not fill the gap for a precise national distribution and independent employer quotes.
Return to the current specification to check entry points, prerequisites, outcomes, and failure conditions; an old version, model output, or fluent tone cannot substitute for facts. Screenshot and image use and expert authority are checked separately.
CAI is a US occupational interest clue, not a personal threshold. See whether the person is willing to re-check, ask questions, revise, and meet real working conditions, with no type or test hiring line.
Preserve a one-page quick start, a step-by-step source matrix, the five revisions from v1 to v2, and the three-path paper walkthrough reflection; mark the fictional input, your own contribution, and that no real testing was done.
Submit to the product or domain technical lead for confirmation; authors keep the original basis and questions. Models cannot invent recovery permissions or substitute for engineering safety or release approval.
FermatMind career content editorial
2026-10-01 original production and first reading; 2026-10-02 only the EP and NBS new review supplement, and historical legal and wage periods are each preserved.
This round newly checked the BLS27-3042 overall EP and the NBS scientific research and technical services two-unit pre-tax scope, with the fourth row's actual statistical link corresponding to the source EP. The original O*NET and four adjacent occupations, 2025 wages, OOH outlook explanation, copyright and personal information, and NIST remain within their original reading scope, and not all regulations were re-read. The notebook v2 specification's four works are a fictional paper walkthrough, not real software or user outcomes.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Not China wages, personal hiring probability, or measured AI accuracy and displacement numbers.
Usage limits：General copyright rules are not technical writing qualifications or wages; employment work ownership and public use require checking the actual agreement.
Usage limits：Does not extrapolate a unified China qualification; the older 2024–2034 employment projection is not used as the current value.
Usage limits：Not China wages, personal hiring probability, or measured AI accuracy and displacement numbers.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：The readable search definitions are not a wage sample for this occupation; averages are not used to infer personal income.
Usage limits：Does not prove technical writing adoption rates, time savings, wage changes, or measured documentation outcomes.
Usage limits：Percentiles are not seniority grades or personal take-home pay; annual values use published values and are not annualized independently.
Usage limits：Does not extrapolate a unified China qualification; the older 2024–2034 employment projection is not used as the current value.
Usage limits：Does not prove technical writing adoption rates, time savings, wage changes, or measured documentation outcomes.
Usage limits：Percentiles are not seniority grades or personal take-home pay; annual values use published values and are not annualized independently.
Usage limits：The official 400 cannot be replaced by subtracting rounded endpoints to get 300; this is not a China projection or personal hiring probability.
Usage limits：Not China wages, personal hiring probability, or measured AI accuracy and displacement numbers.
Usage limits：Does not prove technical writing adoption rates, time savings, wage changes, or measured documentation outcomes.
Usage limits：Not all processing depends solely on consent; access does not automatically mean it can be disclosed or uploaded to models.
Usage limits：Does not prove technical writing adoption rates, time savings, wage changes, or measured documentation outcomes.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Does not extrapolate a unified China qualification; the older 2024–2034 employment projection is not used as the current value.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Does not prove China qualifications or that you can transfer directly; this page does not use that occupation's wages.
Usage limits：Not China wages, personal hiring probability, or measured AI accuracy and displacement numbers.
Usage limits：Percentiles are not seniority grades or personal take-home pay; annual values use published values and are not annualized independently.
Usage limits：Does not prove technical writing adoption rates, time savings, wage changes, or measured documentation outcomes.
Usage limits：Not all processing depends solely on consent; access does not automatically mean it can be disclosed or uploaded to models.
Usage limits：Not external statistics, enterprise recruitment, statutory qualifications, or actual software verification.
Usage limits：Conditional industry background, not precise technical writer wages or recruitment quotes
For career exploration, not substituting for specific post requirements, technical approval, or contracts. If China wages, representative recruitment, and AI causality data were not obtained, the gaps are publicly preserved. Statistics, regions, and contracts are explained separately; US wages are not used to fill the China gap. The notebook v2 is a fictional paper walkthrough only and does not claim real product and user testing. Image, text, and log use, technical facts, and release responsibility are checked separately, and unknowns remain as questions.
MBTI (English) · Big Five (English) · MBTI (Chinese) · Big Five (Chinese)
Check copyright licensing, employment terms, and the basis for personal information processing separately; being able to read customer screenshots does not automatically permit publishing them or entering them into third-party models.
Public works must check image and text sources, employment or commission terms, and log and screenshot use. Personal information processing has several lawful bases; account access does not mean it can be disclosed or uploaded to public models.
Next step：Add complex change impact analysis.
Typical duties：Coordinate continuous update responsibility and content review, and do not treat drafts as published by default.
Promotion evidence：Able to explain the real modification and responsibility handoff process.
Qualification boundaries：Process experience does not equal enterprise document system permissions or record destruction qualifications.
Next step：Continue documentation maintenance or separately check document management duties.
Qualification boundaries：Author scope, expert scope, and release roles are checked separately.
Next step：Add complex product knowledge under real feedback.
Typical duties：Improve instruction accuracy and maintenance ability through repeated real project feedback.
Promotion evidence：List only tasks you actually did, and do not exaggerate experience by naming customers.
Qualification boundaries：Moving to editing, management, or a specialty technical post requires independently checking qualifications and ability.
Next step：Deepen documentation or add outputs for the target occupation.
Usage limits：Editorial ratings use the existing occupational responsibilities and AI task analysis on this page. They are not adoption rates, measured effects, job-disappearance or personal-unemployment probabilities, or official scores from Stanford or another external institution. No measured time weights or future robot deployment are assumed.
