Machine Learning Engineer Jobs: What You Do, Which AI Degree Fits, and How to Start

A machine learning engineer builds models that let software predict, sort or recommend from data, then keeps them working in real products. BLS has no occupation by that name, so we use its data scientist group, which had a May 2025 median wage of $120,230. Most people reach the job with a bachelor’s or an AI master’s.

What Does a Machine Learning Engineer Do?

BLS counts machine learning engineers with data scientists, so these figures cover that wider group rather than machine learning engineers alone. Their duties are the closest official description of the work.

• Determine which data are available and useful for the project.

• Collect, categorize and analyze that data.

• Create, validate, test and update algorithms and models.

• Use data visualization software to present findings.

• Make business recommendations to stakeholders based on the analysis.

Here is how that looks on a Tuesday. You pull a data set, clean it, and train a model. The model misses, so you look at why, change the features or the training setup, and run it again. When it finally works, you check it against data it has never seen, and you write down what it can and cannot do.

Then comes the part many students skip. A model is not finished when the score looks good. Someone has to run it inside an app, watch it drift as the world changes, and explain a surprising answer to a manager. That is why employers want people who can both do the math and write software.

Which Job Titles Should You Search?

Employers do not agree on a title, so searching one phrase will miss good openings. Try machine learning engineer, ML engineer, data scientist, applied scientist and AI engineer, and read the duties rather than the label.

Our AI careers overview shows how the research, software and data scientist titles split. In short, research scientists lean toward new methods, software developers toward building the product, and data scientists toward analysis and models. A machine learning engineer sits somewhere across all three, depending on the company.

For pay on the software side, see our AI engineer salary page. Its figures come from a different BLS group, so the two pages will not match.

Which AI Degree Level Do Employers Expect?

BLS lists a bachelor’s degree as the typical entry education for data scientists, and it says graduate school is not required. That is the federal definition of typical, not a hiring rule, and individual employers set their own bar.

Our program data shows how the options stack up.

• Bachelor’s degrees. Online AI bachelor’s programs exist, including Dakota State University at $44,856 total, Grand Canyon University at $58,800, University of Maryland Global Campus at $59,880 and the University of Tennessee-Knoxville at $67,800. Western Governors University’s B.S. in AI Engineering charges $4,200 per six-month term, so you pay less if you finish sooner.

• Master’s degrees. This is where machine learning gets its own courses, and where most of our program data sits.

• Doctorate. Capitol Technology University’s online PhD in Machine Learning costs $58,800 and is built for working professionals who already have experience.

A bachelor’s can get you in the door for some roles. A master’s is the level where you study machine learning, deep learning and language models in depth, which is why many people pursue it after a computer science or math undergraduate degree. Our AI degree vs computer science degree guide covers that trade.

Why Do Many Machine Learning Engineers Get a Master’s?

The honest reason is depth. A four-year degree usually gives you programming, statistics and a survey of AI. A master’s gives you time to train real models, read the research, and finish a capstone. Whether a given employer asks for it is something to check in its postings.

We cannot prove it moves your pay, because BLS does not report wages by degree for this occupation. What we can show is what each kind of program teaches.

The University of Texas at Austin’s online master’s lists Machine Learning, Deep Learning, Natural Language Processing and Reinforcement Learning, with Ethics in AI required. DePaul lists Applied Deep Learning, Natural Language Processing and Machine Learning I. City University of Seattle includes cloud computing and agent-based systems alongside machine learning and deep learning.

If you are comparing programs, start with our online machine learning degrees ranking. It filters for programs that teach machine learning at the master’s and doctoral levels.

Which Program Types Fit Machine Learning Engineer Jobs?

Not every AI degree prepares you for model work. The names tell you less than the curriculum, so match the type to the job you want.

Machine learning specializations. Purdue’s online MS in AI offers an AI and Machine Learning major. West Virginia University offers Machine Learning and Intelligent Systems concentrations. The University of Michigan-Dearborn and the University of North Texas each offer a Machine Learning concentration. George Washington University and Drexel put machine learning in the program name. The machine learning degrees ranking lists them side by side.

AI engineering programs. These lean toward building and deploying systems. Penn State World Campus offers an AI Engineering option, Duke’s is a Master of Engineering, and Boston University has a software engineering for AI master’s. See the AI engineering degrees ranking.

General AI master’s programs. Broader and often cheaper, they cover machine learning as one part. Browse them in our online master’s in AI ranking.

Programs without the GRE. Several of the master’s programs above waive or do not require the GRE. Our no-GRE master’s list shows which.

What Do You Need Before You Apply?

Prerequisites decide whether you can start now or need a bridge year.

• West Virginia University wants a bachelor’s in computer science, computer engineering, cybersecurity or a closely related field with a 3.0 GPA. Other fields may be considered provisionally with a year of AI or machine learning experience.

• Purdue University wants a 3.0 GPA, and its machine learning major needs programming experience plus coursework in algorithms, calculus, linear algebra and probability.

• University of Michigan-Dearborn asks for a STEM bachelor’s with a 3.0 GPA and prior programming, math and statistics coursework.

• DePaul University asks for a 2.5 GPA and Calculus 1 and 2, and says programming experience strengthens an application.

• Drexel University requires a strong computer science background and points to its Graduate Certificate in Computer Science as an entry point.

• University of Colorado Boulder admits on performance: you pass a three-course pathway with a B or better, and no bachelor’s degree is required.

If you lack the math, plan for a semester or two of catching up. That is a real cost in time, and it is better to know now than in week three.

What Skills and Tools Do Machine Learning Engineers Use?

BLS names the skills in the duties: finding useful data, analyzing it, building and testing models, and presenting the results. School programs fill in the rest.

• Math. Calculus, linear algebra, probability and statistics, the same subjects Purdue and DePaul list as prerequisites.

• Programming. Every program on our list expects it, though how much varies.

• Model building. Supervised learning, deep learning, language models and reinforcement learning appear in course lists from Texas, Penn and DePaul.

• Cloud and systems. City University of Seattle teaches AI in the cloud, and engineering programs cover deployment.

• Communication. BLS lists presenting findings and making business recommendations as core duties.

• Ethics. Texas requires Ethics in AI, and DePaul and Penn include ethics courses.

Tools change faster than degrees do, so read three recent postings for the role you want and note the ones that repeat.

How Do You Get Hired as a Machine Learning Engineer?

There is no single route, but these steps cover most of them.

1. Pick the role first. Decide whether you want to build products, analyze data, or do research. Our guide to becoming an AI engineer explains the split.

2. Check your math and coding. If you can write working code and have taken calculus and linear algebra, you are ready for most programs. If not, take those first.

3. Choose a degree level. A bachelor’s for entry roles, a master’s if you want the machine learning depth, a doctorate only for research jobs.

4. Pick a program with a project. Capstones at Drexel, DePaul and City University of Seattle give you something to show. Drexel, DePaul and CityU also list internships.

5. Build work you can show. A trained model, the data you used, and a short write-up on what failed will tell an employer more than a course list.

6. Search several titles. Use the list above, and read each posting’s duties.

7. Apply in the right places. California has the most jobs and Washington has the highest pay, but plenty of openings sit elsewhere.

What Do Machine Learning Engineers Earn?

For the closest BLS group, the national median is $120,230 a year, or $57.80 an hour. The mean is $126,800. The 10th percentile is $67,240 and the 90th is $199,130, and the middle half earn between $85,660 and $158,880.

By state, the highest mean wages are in Washington at $158,040, California at $156,000, Maryland at $144,080, the District of Columbia at $142,580 and New Jersey at $138,590.

The most jobs are in California with 39,310, Texas with 25,860, New York with 23,970, Pennsylvania with 13,810 and North Carolina with 11,430. Our machine learning engineer salary page has the full breakdown, with percentiles and degree costs.

What Is the Job Outlook?

BLS projects data scientist employment to grow 35 percent from 2025 to 2035, which it calls much faster than average. That is about 95,400 more jobs on a base of 275,600, with about 24,800 openings a year, counting people who leave the occupation as well as new jobs.

Projections are estimates, and BLS revises them. They also say nothing about which candidates get the jobs. Employers may prefer a specific degree or experience, and a posting can close the day it opens.

Think of growth as a reason to try, not a guarantee. A strong portfolio matters more than any percentage.

Can You Do This Job Remotely, and Can You Earn the Degree Online?

The degree is the easier question. Every online program we list for machine learning is fully online, with no required campus visits. University of Washington is the one note: full online completion is possible part time, but some optional certificate components are in person only.

The job is harder to answer. Postings range from fully remote to fully on site, and the mix changes. Read the location line, and ask in the interview.

One more point on cost. Online master’s tuition among the programs we list runs from $9,990 at the University of Texas at Austin to $81,000 at Columbia, so the sticker price is not a reason to skip the comparison. See our salary page for how those prices stack up against the pay.

What Are the Downsides?

Be ready for these before you commit.

• The title is not tracked, so you cannot see federal pay for it. You see a nearby group and need to read postings yourself.

• Prerequisites are real. Several programs expect math and programming you may need to build first.

• Cost varies a lot, and some programs charge per credit before fees, so the final bill can be higher than the tuition line.

• Projections describe an occupation, not your odds. A degree helps, but experience and projects decide many offers.

• The work changes fast. Whatever you learn will need updating every year or two.

None of that should talk you out of it. It should make you pick a program on purpose and plan for the gaps.