Computer Hardware Engineers
In short: today's best AI models could do about 23% of this job's working time (projected from METR's April 2026 measurement and Epoch's November 2026 index) if the work were set up for them, and about 63% by the end of 2028 if progress keeps its long-run pace. The conversations, in-person and hands-on work, about 30% of the time, stays with people.
Research, design, develop, or test computer or computer-related equipment for commercial, industrial, military, or scientific use. May supervise the manufacturing and installation of computer or computer-related equipment and components.
76,660 people do this job in the US. $161,740 a year.
Reviewed on October 1, 2026: what each task is made of, and what handing it over would ask.
A typical 40-hour week
- AI could take on2 h
- A person checks AI's work7 h
- Desk work AI can't do yet19 h
- Stays with people12 h
Computer Hardware Engineers, now: of a 40-hour week, AI could take on 2 hours, 7 hours become checking its work, 19 hours are desk work it can't do yet, and 12 hours stay with people.
The largest pieces: Write functional specifications (about 2 h, a person checks); Handle data for analysis (about 1 h); Determine hardware configuration (about 1 h, a person checks).
Estimates for a typical week of this job, if progress keeps its long-run pace, counting work the best models finish four times in five. Every task is below.
of the working time. Now is projected from METR's April 2026 measurement and Epoch's November 2026 index.
- 66%
- 4%
- 6%
- 3%
- 22%
of the work: Anthropic's measure of what people already bring to Claude at work (2025). It counts help with part of a task, so it can be higher than the share within reach.
On the chosen pace and reliability. Estimate.
Share of the working time within reach, now to 2030
Change assumptionsHide assumptionsLong-run pace, four in five
What each choice means: , , , , . Every pace starts from .
At this date and pace, models finish tasks up to 2.4 work days long four times in five.
What handing it over would take, now
The 23% within reach splits by what a model would need. The dates, pace and reliability follow the controls above. Estimates.
of the working time. The material can be handed to a model and the result handed back.
of the working time. The model has to be connected to the system the work lives in.
is conversation short enough by length. This site leaves conversation with people, so it is not in the total.
How far this has come
With GPT-4, measured in March 2023, 0% of this job's working time was within reach at . With Claude 3.7 Sonnet, measured in February 2025, 0%. With Claude Mythos Preview (early), measured in April 2026, 4%. Today's estimate: 23%.
A backcast: today's task data with the best model METR had measured by each date. It shows how fast the models have moved, not what anyone forecast at the time.
Measured use over time
Of the Claude.ai conversations Anthropic sampled and tied to a work task, 0.87% went to this job's tasks in May 2026, the 31st largest of the 447 jobs that showed up at all; 4 of its 18 tasks showed up. Within each of Anthropic's methods, never across a change: 0.59% to 0.67% (released February and March 2025); 0.27% to 0.36% (August 2025 to February 2026); 0.06% to 0.87% (April to May 2026).
Each release
| Conversations | Share | Rank | Tasks seen |
|---|---|---|---|
| Released February 2025 | 0.59% | 29 of 708 | 7 of 18 |
| Released March 2025 | 0.67% | 28 of 703 | 5 of 18 |
| The week of Aug 4, 2025 (released September 2025) | 0.27% | 67 of 655 | 6 of 18 |
| The week of Nov 13, 2025 (released January 2026) | 0.33% | 52 of 672 | 9 of 18 |
| The week of Feb 5, 2026 (released March 2026) | 0.36% | 60 of 677 | 8 of 18 |
| April 2026 (released June 2026) | 0.06% | 212 of 434 | 3 of 18 |
| May 2026 (released June 2026) | 0.87% | 31 of 447 | 4 of 18 |
Measured, not estimated: Anthropic's samples of Claude.ai conversations, each tied by a model to an O*NET task, from the Anthropic Economic Index. It says where Claude's use goes, not how much of this job's working time it covers, so it is not on the same scale as the reach above. The releases differ in their sample, their classifier and the smallest share they report (the June 2026 file rounds each task to a hundredth of a percent, so a smaller task reads as none), which is why a change of method starts a new line. At least 96% of each older release's task-tied use matched this site's tasks, and all of the newest. Claude.ai only, so use through other tools and other companies' models is not in it. Every job ranked.
Robots and the physical work
Anthropic rates 29% of this job's working time as physical work. For 27% of the working time, a robot has been shown doing most of the task somewhere: 27% in a purpose-built cell. 0.27 of 3, the 399th highest of the 923 jobs.
- Not physical 71%
- No robot shown 3%
- In a purpose-built cell 27%
Anthropic's rating, not this site's estimate: Claude Opus 5, searching the web, rated each physical O*NET task by the kind of setting in which a robot it could cite (deployed, sold or demonstrated) does most of the task, weighted by its own estimate of each task's share of the time (What work can robots do?, September 30, 2026, data CC BY 4.0; every job ranked). A robot shown somewhere is not a robot in your building: cost, setup and safety decide that. The share within reach on this page still counts no body work. The robot behind each rated task is under the task below.
What changes for the people in this job
From the task-by-task review, by the end of 2028 on the long-run pace at four in five. It says which parts of the work stay with people and which change, not whether anyone keeps a job: that depends on decisions no data set can see.
What stays with people
30% of the working time: the conversation, in-person and hands-on parts of the tasks.
- Build and test prototypes, the hands-on work, 11%
- Test and verify hardware, the hands-on work, 6%
- Monitor and adjust equipment, the hands-on work, 4%
- Keep skills current, the in-person work, 2%
- Direct technical staff, the conversation, 2%
- Confer on hardware and software interfaces, the conversation, 2%
What becomes checking the machine
57% of the working time: desk work a model could do by then, in tasks where a mistake costs something or a licensed person signs. A person checks it. Another 6% is desk work that is small if wrong, which can be handed over with lighter checks.
- Design computer hardware, costly if wrong, 16%
- Write functional specifications, costly if wrong, 12%
- Build and test prototypes, costly if wrong, 10%
- Test and verify hardware, costly if wrong, 5%
- Determine hardware configuration, costly if wrong, 4%
- Monitor and adjust equipment, costly if wrong, 3%
What to learn
- Briefing a model and checking its drafts. About 66% of the time is documents and data, the part models reach first: writing a clear brief, giving the right material, and checking what comes back.
- Reviewing work you did not do yourself. The tasks under What becomes checking the machine, where a mistake costs something, stay a person's to check: spotting what a draft gets wrong is a skill of its own.
- The hands-on craft. About 22% of the time is hands-on work, which no model here does; robots are rated separately on this page.
Your own version of this job: read it against your job description. The headline is free; the full personal report, every task with its timeline and what to learn, is $19 (a sample).
The tasks
Sorted by . 18 tasks from .
| Task | ||||
|---|---|---|---|---|
| Design and develop computer hardware and support peripherals, including central processing units (CPUs), support logic, microprocessors, custom integrated circuits, and printers and disk drives.Design work on the computer.Done more than yearly by most who do it. | 22% | 4.2 work days6.7 h to 4.2 work weeks | Documents 100% | 5%Half by Jun 2028 |
| Build, test, and modify product prototypes, using working models or theoretical models constructed with computer simulation.Prototypes at the bench and in simulation.Done more than weekly by most who do it. Robots: in a purpose-built cell, Tempo Automation smart factory (automated prototype PCBA line) (sold). Anthropic's rating. | 21% | 2.1 work days3.3 h to 1.7 work weeks | Documents 50%, hands-on 50% | 7%Desk part half by Oct 2027 |
| Write detailed functional specifications that document the hardware development process and support hardware introduction.Writing the specifications.Done more than yearly by most who do it. AI observed on this task. | 12% | 1.7 work days3.3 h to 4.2 work days | Documents 100% | 31%Half by Mar 2027 |
| Test and verify hardware and support peripherals to ensure that they meet specifications and requirements, by recording and analyzing test data.Running the tests and analyzing the data.Done more than monthly by most who do it. Robots: in a purpose-built cell, Advantest M4171 test handler (sold). Anthropic's rating. | 11% | 2.1 work days3.3 h to 1.7 work weeks | Documents 50%, hands-on 50% | 7%Desk part half by Oct 2027 |
| Monitor functioning of equipment and make necessary modifications to ensure system operates in conformance with specifications.Watching the system and making changes at the bench.Done daily by most who do it. Robots: no robot shown. Anthropic's rating. | 7% | 3.3 h50 min to 2.1 work days | Hands-on 60%, systems 40% | 25%Desk part half by now |
| Store, retrieve, and manipulate data for analysis of system capabilities and requirements.Pulling the data and working it.Done more than weekly by most who do it. AI observed on this task. | 4% | 2.5 h50 min to 1 work day | Documents 70%, systems 30% | 92%Half by now |
| Update knowledge and skills to keep up with rapid advancements in computer technology.Reading and courses; the knowing is the engineer's own.Done more than monthly by most who do it. | 4% | 2.5 h50 min to 6.7 h | Documents 50%, in person 50% | 48%Desk part half by now |
| Evaluate factors such as reporting formats required, cost constraints, and need for security restrictions to determine hardware configuration.Weighing the requirements and deciding the configuration.Done more than monthly by most who do it. | 4% | 5 h1.7 h to 2.1 work days | Documents 100% | 73%Half by now |
| Direct technicians, engineering designers or other technical support personnel as needed.Handing out work and reviewing it, in conversation and on paper.Done more than weekly by most who do it. | 3% | 1.3 h25 min to 3.3 h | Conversation 60%, documents 40% | 40%Desk part half by now |
| Confer with engineering staff and consult specifications to evaluate interface between hardware and software and operational and performance requirements of overall system.Working meetings with the specifications in view.Done more than weekly by most who do it. | 3% | 1.7 h38 min to 5 h | Conversation 60%, documents 40% | 39%Desk part half by now |
| Analyze user needs and recommend appropriate hardware.Finding out what users need and writing the recommendation.Done more than monthly by most who do it. AI observed on this task. | 2% | 2.5 h50 min to 1 work day | Documents 70%, conversation 30% | 64%Desk part half by now |
| Provide technical support to designers, marketing and sales departments, suppliers, engineers and other team members throughout the product development and implementation process.Answering questions by call and in writing, some from suppliers.Done more than monthly by most who do it. | 2% | 1.7 h25 min to 6.7 h | Conversation 60%, documents 40% | 35%Desk part half by now |
Within reach, per task: the share of the task's time whose instances are short enough for a model to finish at the chosen date, pace and reliability. Half by: the date half of that time comes within reach. Made of: how the task's time splits among five kinds of doing. Only the documents and systems parts can count as within reach, so a task that is 30% documents can be at most 30% within reach.
Jobs that share skills with this one
From O*NET's related occupations. Within reach now and by the end of 2028, long-run pace, four in five.
- Electronics Engineers, Except Computer26% now, 73% by 2028
- Electrical and Electronic Engineering Technologists and Technicians39% now, 48% by 2028
- Microsystems Engineers10% now, 63% by 2028
- Software Developers48% now, 78% by 2028
- Robotics Technicians22% now, 26% by 2028
- Mechatronics Engineers19% now, 74% by 2028
Where this job works
The industry groups that employ the most of it (BLS, May 2025).
Where the most of these jobs are
The metro areas with the most people in it (BLS, May 2025).
Questions about this job
- How much of the work of Computer Hardware Engineers is within reach of AI now?
- An estimated 23% of the working time is within reach of AI models now and 63% by the end of 2028, on METR's long-run pace at four in five (47% to 69% across the three paces the site shows). Within reach means a model could finish work that long at that reliability. It is not a forecast of jobs lost or of savings.
- Which tasks of Computer Hardware Engineers come within reach first?
- Weighted by the time each takes, the tasks with the most within reach now (projected from METR's April 2026 measurement and Epoch's November 2026 index) are: Write functional specifications (31% of the task's time now, 100% by the end of 2028); Handle data for analysis (92% of the task's time now, 100% by the end of 2028); and Determine hardware configuration (73% of the task's time now, 100% by the end of 2028).
- Which tasks of Computer Hardware Engineers stay with people?
- Tasks that are people work or body work are not counted as within reach at any date. The largest: Train designers and users; and Modify equipment. Together such tasks are about 2% of the working time, and the people and body parts of the other tasks stay with people too.
- What should Computer Hardware Engineers learn as AI takes on more of the work?
- By the job's own task-by-task review, the first things to learn are briefing a model and checking its drafts; reviewing work you did not do yourself; and the hands-on craft. By the end of 2028 on the long-run pace at four in five, about 30% of the working time stays with people, 57% becomes checking a model's work and 6% could be handed over with lighter checks. It says which parts of the work change, never whether anyone keeps a job.
- Do people already use AI for the work of Computer Hardware Engineers?
- In the Claude.ai conversations Anthropic sampled for May 2026 and tied to a work task, 0.87% went to this job's tasks, the 31st largest share of the 447 jobs that showed up; 4 of its 18 tasks appeared. Anthropic's labor market measure puts the share of the job's tasks seen in real Claude use at work at 15%. These are measures of use, not of how much of the work AI could finish, and they cover Claude only.
- Can robots do the physical work of Computer Hardware Engineers?
- Anthropic's robot exposure rating (September 2026) finds 29% of the working time is physical work, and that a robot has been shown doing most of the task, somewhere, for 27% of the working time. Its index is 0.27 of 3. One example: Tempo Automation smart factory (automated prototype PCBA line) for "build and test prototypes". A robot shown somewhere is not one in every workplace; this site's own estimates still count no physical work as within reach.
- How is this estimated?
- From the 18 tasks O*NET lists for the occupation, each with an estimated length checked against GDPval and a share of the working time, reviewed one task at a time on October 1, 2026 for what each is made of (documents, business systems, conversation, in person, hands-on), and METR's measurements of how long a task the best AI models can finish, carried forward to the date from the latest, Claude Mythos Preview (early) in April 2026. Data checked November 10, 2026.
This job in your state or metro area
Embed these figures on your site
A small card with the share of working time within reach now and at the end of 2027, 2028 and 2030, the date and a link back. Free under CC BY 4.0; the card carries the credit.
Your own version of this job
The page above is the occupation. Your job is your title and what you actually do: give both, and they are read against these tasks in your browser, nothing sent anywhere. The headline is free: how much of your working time is within reach now and by the end of 2028, and how many of the tasks your description names.
The full personal report is $19, to print or save as a PDF: every task with when it comes within reach, which ones your description names, the parts of your description the task list does not cover, what stays with you, what becomes checking the machine and what to learn.