How much do AI engineers make a year?
Five sources, five different numbers. Each is that source's own report, and each counts something slightly different.

| Source | What it measures | Figure | As of |
|---|---|---|---|
| Levels.fyi, AI Engineer | Median total compensation, 300+ salaries | $155,500 | Updated 21 November 2026 |
| Levels.fyi, ML / AI software engineer | Median total compensation | $251,250 | Updated 22 November 2026 |
| Built In, AI Engineer | Average base salary; average total $211,243 | $184,757 | Titled 2026, no date stated |
| Glassdoor, as reported by Coursera | Median base pay, AI engineer | $134,188 | Glassdoor data as of February 2025 |
| BLS, computer and information research scientists | Median annual pay, a related occupation | $140,300 | May 2025 |
No single row is wrong. Base pay leaves out bonus and stock, and total pay adds them. A title like "ML / AI software engineer" draws more reports from large tech companies, which pay more than the average employer.
Why do AI engineer salary reports disagree?
Four things move the number, and a reader can check each one.
- Base versus total. Built In's own page shows a $184,757 average base and a $211,243 average total, a $26,486 gap in cash alone, before any stock.
- Job title. Levels.fyi files ML / AI software engineers apart from AI engineers and reports $95,750 more for the first group.
- Who reports. These sites rely on people who choose to submit a figure. A pool that skews to big tech employers will read higher than one that does not.
- Age. Coursera's February 2026 guide cites Glassdoor figures from February 2025, and Coursera's May 2026 guide gives a different Glassdoor median ($134,023).
The Reddit thread that ranks first for this query, titled "24M AI Engineer making 530k", is a single anonymous post. It shows what the top of the market can look like. It gives no median, no sample and no way to verify the number, so treat it as one data point.
What is the entry level AI engineer salary?
Glassdoor's base pay for an AI engineer with under one year of experience is $103,140, as Coursera reported it in February 2026. Built In gives a lower $80,000 for under one year, so the first figure depends on whose data you read.
| Experience | AI engineer base pay (Glassdoor, via Coursera) | Machine learning engineer base pay (Glassdoor, via Coursera) |
|---|---|---|
| 0 to 1 years | $103,140 | $98,798 |
| 1 to 3 years | $121,641 | $112,105 |
| 4 to 6 years | $138,301 | $122,505 |
| 7 to 9 years | $155,132 | $133,130 |
| 10 to 14 years | $172,468 | $153,286 |
The curve is steady rather than steep: Glassdoor's own AI engineer figures rise about $69,000 from the first year to the 10 to 14 year band. Built In reports $194,413 for people with 7 or more years, which is above Glassdoor's $155,132 for the same band. Different pools of reports produce different levels, but both show pay rising with experience.
How much does a machine learning engineer make?
Glassdoor reports a median base pay of $123,117 for machine learning engineers, as Coursera listed it in February 2026, against $134,188 for AI engineers. On Levels.fyi the ML / AI software engineer group reports a $251,250 median in total pay, with a 25th percentile of $170,000, a 75th percentile of $375,500 and a 90th percentile of $517,555 (updated 22 November 2026).
The two titles overlap heavily in job posts. Glassdoor puts machine learning engineers below AI engineers on base pay at every experience band in the table above, while Levels.fyi's separate ML / AI group sits well above its AI engineer median. The honest reading: the title matters less than the employer and whether the figure includes stock. What is machine learning covers what the role actually builds.
Does location change an AI engineer's pay?
Yes, and San Francisco leads in both sources that list cities. Glassdoor via Coursera and Built In each rank it first, with New York second.
| City | Glassdoor AI engineer base (via Coursera) | Glassdoor ML engineer base (via Coursera) | Built In AI engineer average |
|---|---|---|---|
| San Francisco, CA | $164,499 | $149,795 | $246,250 |
| New York City, NY | $141,539 | $124,920 | $223,464 |
| Houston, TX | $129,169 | $121,462 | not stated |
| Chicago, IL | $106,031 | $110,268 | $177,300 |
| Seattle, WA | not stated | not stated | $180,000 |
| Boston, MA | not stated | not stated | $172,500 |
| Remote | not stated | not stated | $180,173 |
Built In's cities show an average salary and Glassdoor's show base pay, so compare down a column and not across. One oddity: Glassdoor's Chicago figure for machine learning engineers ($110,268) is above its AI engineer figure ($106,031), the one city where the order flips.
Which employers pay AI engineers the most?
The Bureau of Labor Statistics doesn't publish a median for AI engineers as a job title. Its closest category is computer and information research scientists, and it does publish pay by industry for them in May 2025.
| Industry (BLS, May 2025) | Median annual pay |
|---|---|
| Miscellaneous information services | $218,560 |
| Computer systems design and related services | $162,760 |
| Research and development in the physical, engineering and life sciences | $141,830 |
| Federal government, excluding postal service | $124,870 |
| Colleges, universities and professional schools; state | $85,460 |
The spread is $133,100 from top to bottom. Information services, which include large internet companies, pay about 2.6 times the state university median. Levels.fyi publishes no company type table for this title. Its page lists only individual reports, covered next, so any ranking of frontier labs against banks against startups would be a guess.
What does Indeed report for AI engineer pay by city?
Coursera's 2026 guide relays Indeed annual base salary figures for the AI engineer title in eight cities. Indeed's own page was blocked to us, so these are Indeed numbers as Coursera printed them, with no collection date given.
| City | Indeed annual base (via Coursera) |
|---|---|
| San Jose, CA | $206,706 |
| Boston, MA | $189,318 |
| New York, NY | $189,274 |
| Santa Clara, CA | $186,734 |
| Nashville, TN | $158,077 |
| Dallas, TX | $151,692 |
| Austin, TX | $138,701 |
| Seattle, WA | $136,347 |
Seattle sits at $180,000 on Built In and $136,347 on Indeed, a $43,653 gap for one city. Same lesson again. The pool of reports decides the number, so compare down one source and not across.
How can an AI engineer earn more?
Pay reports point to three levers, and each one shows up in the tables above. Experience is the steadiest: Glassdoor's AI engineer base rises from $103,140 to $172,468 across its bands. City is the next, and San Francisco leads every source that lists cities, though the size of its lead over New York changes from one source to the next, so check which pool of reports you are reading before you plan around it. Counting stock and bonus is the third, since Built In adds $26,486 in cash on top of base before any equity.
None of these is a promise. Pay is set by one employer at a time. Ask for base, bonus and stock in writing and check the offer against two sources.
Do AI engineers make more than software engineers?
On the BLS numbers, barely, and on Levels.fyi's, noticeably. BLS reports a $134,040 median for software developers in May 2025 and $140,300 for computer and information research scientists, a gap of $6,260. BLS lists a bachelor's degree as typical entry education for the first and a master's for the second.
The outlook figures differ more. BLS projects 10% growth from 2025 to 2035 for software developers, about 1,905,400 jobs in 2025, and 22% for research scientists, about 38,600 jobs. Those are BLS projections, and it is a small occupation, so a few hundred hires move the percentage. AI job loss statistics covers the other side of that story.
How do you become an AI engineer?
Most paths start with a software engineering base, then add model skills. The five steps below match what job posts and the BLS entry education lines describe.
The work itself explains the pay. An AI engineer builds products on top of models: calling them, giving them the right documents, testing the answers and keeping the cost per request in check. That is closer to software engineering than to research, which is why the first step is the same as for any developer.
A degree isn't the only route, but both BLS occupations above list one as typical. Coursera's guide cites Zippia for the education split among AI specialists: about 63% hold a bachelor's and about 17% a master's. That is Zippia's figure for a related job title, relayed by Coursera, so don't read it as a requirement.
Use prompt engineering for beginners as a first project, and AI training jobs if you want a nearer way in.
What can an AI engineer salary report not tell you?
It can't tell you what you will be paid. A report is a sample of people who chose to share a figure, and it says nothing about your city, your employer or your negotiation. This page is a reading of other sites' data, not career or financial advice.
Every figure here is a snapshot, and Levels.fyi's says it was updated on the day this page was written. The projections are BLS's, and ours would be a guess. For a real offer, ask the employer for the base, bonus and stock figures in writing, and check them against two sources.
What can you ask an AI about AI engineer pay?
A model can help you read pay data, but only if you give it the data. These five prompts work in Claude, ChatGPT, Gemini or Whizi, with web search on.
- Split an offer: "My offer says [number] for an AI engineer in [city]. Ask me whether it is base or total pay, then list what a base only report and a total pay report would each show."
- Check a source: "Open [page] and tell me the date of the figure, whether it is base or total pay and how many reports it rests on. Say not stated for anything the page doesn't say."
- Compare titles: "Compare AI engineer and machine learning engineer pay for [city] using only Levels.fyi, Glassdoor and BLS. Make a table and label each cell with its source and date."
- Spot the outlier: "Here are five salary reports: [paste]. Which one doesn't fit the others and what might explain it?"
- Plan the skills: "I know [languages and tools]. List the three biggest gaps between me and an entry level AI engineer job post, with one small project for each."
Ask for sources and open them. A model can mix up base and total pay as easily as a person can.
Where does Whizi fit?
Whizi is one app instead of a pile of single purpose tools, and it suits you if you want to read pay data and test your questions on more than one model. Web search is on by default in each chat, and Compare, which runs two models on the same prompt side by side, is on Pro. The free plan gives you 10 messages a day on Whizi AI, Auto and Claude Haiku, with no card, and Pro starts with $1.99.
- Check whether a figure is base pay or total pay before you compare it
- Note the title: AI engineer and ML / AI software engineer report different numbers
- Look for the date and the number of reports behind each figure
- Compare cities down one source's column, never across two sources
- Ask the employer for base, bonus and stock in writing
Frequently asked questions
How much does an AI engineer make in the US?
Levels.fyi reports a $155,500 median total pay for AI engineers, as of 21 November 2026. Glassdoor reports a $134,188 median base, as Coursera listed it in February 2026, and Built In reports a $184,757 average base. The range comes from base versus total pay.
Is a machine learning engineer paid more than an AI engineer?
It depends on the source. Glassdoor reports a lower median base for machine learning engineers ($123,117) than for AI engineers ($134,188), while Levels.fyi reports a higher median total for its ML / AI software engineer group ($251,250). The job title matters less than the employer.
What is the highest AI engineer salary?
Levels.fyi reports a 90th percentile of $280,000 for AI engineers and $517,555 for ML / AI software engineers, as of 22 November 2026. One Reddit post claims $530,000, but it is a single anonymous report that can't be checked.