Machine Learning Engineer Salary Guide

Machine learning engineer salary ranges from roughly $128,000 to $186,000 in base pay depending on which aggregator you check, with total compensation at senior levels regularly exceeding $350,000 when stock and bonuses are factored in. The Bureau of Labor Statistics projects 20% growth for the nearest equivalent occupation through 2034, and data scientist roles, which overlap heavily, are growing at 34%. Both numbers are well above the national average for all occupations.

Five different salary databases. Five different numbers. That’s the first thing you notice when you start pulling ML engineer compensation data, and it’s the thing that trips up both candidates and hiring managers more than anything else we see in this space. Glassdoor says $160,347. Indeed says $186,447. ZipRecruiter says $128,769. Built In lands at $162,080. None of them are wrong, exactly. They’re measuring different populations, weighting self-reported data differently, and defining “machine learning engineer” with different boundaries. A candidate who anchors on the Indeed number walks into a negotiation expecting something very different than a candidate who anchored on ZipRecruiter, and we’ve watched that disconnect blow up offers.

We place ML engineers through our AI and ML staffing practice at KORE1, mostly across Southern California but increasingly for remote roles with companies headquartered on the West Coast. The salary conversations in this discipline are harder than most because the gap between base salary and total comp is enormous at the senior end, and because “ML engineer” means different things at different companies. Someone building recommendation systems at a Series D startup and someone fine-tuning foundation models at Meta are both called ML engineers. They’re not doing the same job. They’re not paid the same either.

What ML Engineers Earn Across Five Salary Sources

Every number below is pulled from a named source with a publication date. We’re including the spread on purpose. The variance tells you something about the role’s fragmentation.

The Levels.fyi number jumps off the page. $261,683 median. That’s total compensation, not base, and it skews heavily toward big tech companies where equity packages are a real part of the pay structure. The base salary component on Levels.fyi is $190,000 at the median, which sits comfortably in the range of the other four sources. The delta is stock grants.

Built In reports average total compensation of $212,022 when bonuses and equity are included, based on their own survey data. So even outside the Levels.fyi FAANG bubble, ML engineers are pulling $50,000+ above base in variable comp. That matters for hiring managers writing job descriptions. If your posted range stops at $165,000 and doesn’t mention equity or bonus structure, you’re going to lose candidates to companies that lead with total comp numbers.

Salary by Experience Level

The experience curve is steep. Steeper than software engineering generally, because the supply of people who can actually build production ML systems is still thin relative to demand.

Notice ZipRecruiter’s entry-level number. $69,362. That’s not a typo. ZipRecruiter counts every posting that mentions “machine learning” in the description, including roles that are really data analyst positions with an ML keyword stuffed in for visibility. When we see a candidate with a genuine ML background interviewing for a role at that comp level, something’s off with the job description, not the market.

The 0-to-5-year jump is where the real acceleration happens. Going from $120K to $159K in three years isn’t unusual. But the jump from mid-level to senior is where the negotiation dynamics change completely. A senior ML engineer with production deployment experience, not just Jupyter notebook prototyping, can command $190K+ base without pushing hard. The supply constraint at that level is real. We ran a search last quarter for a client in Orange County who needed a senior ML engineer with PyTorch production experience and Kubernetes fluency. Took 11 weeks. The candidate pool for that specific combination, in SoCal, willing to go hybrid three days a week? Fourteen people. We talked to nine of them. Six were already in counter-offer situations.

Where Geography Moves the Number

California pays more. Not a surprise. But the spread is wider than most people assume.

The California statewide number from Indeed, $210,770, includes the San Francisco gravitational pull. If you strip out the Bay Area and look at Southern California specifically, our placement data shows ML engineer base offers in the $165,000 to $195,000 range for mid-to-senior roles in Orange County, Irvine, and greater LA. Check our 2026 SoCal IT Staffing Salary Report for broader context on how these numbers fit into the regional tech compensation picture.

Remote roles averaging $195,475 is the number that should grab hiring managers’ attention. That’s 21% above the national average, according to Built In. The remote premium for ML specifically has held steady because the talent pool is concentrated in high-cost metros. A company in Raleigh posting a remote ML role still competes with Bay Area and LA employers. The labor market doesn’t care where your headquarters is. It cares where the candidates live.

One thing we’ve noticed in the last six months. Companies that tried to enforce return-to-office mandates for ML roles lost candidates at a higher rate than for other engineering disciplines. An ML engineer who knows PyTorch, has production deployment experience, and can architect a training pipeline doesn’t need to sit in your office to do that work. The ones who are willing to commute can demand a premium for it, and they do.

Total Compensation at Major Tech Companies

Base salary is the floor. At the companies paying top dollar for ML talent, equity and bonuses can double or triple the base number. Levels.fyi tracks self-reported total compensation packages, and the numbers for ML engineers at big tech are striking.

Snap at $450K median. I had to double-check that one. Their ML team is smaller than Meta’s or Google’s, so the data set is thinner, but Snap has been aggressive on ML compensation because they’re competing for the same recommendation-systems talent as companies with ten times the headcount. The range topping $800K at Meta reflects staff and principal-level engineers with multi-year equity vesting.

For context outside big tech: mid-market companies, Series B through D startups, healthcare tech firms, and financial services companies are typically paying $180,000 to $260,000 in total comp for senior ML engineers. The equity component is less liquid, so candidates discount it more heavily. We’ve had candidates turn down startup offers at $230K total comp to take a big tech offer at $340K because the startup equity was “probably worth something, maybe, if the company exits in four years.” Can’t blame them. Stock in a pre-IPO company is a bet. RSUs at Google are cash with a four-year delay.

ML Engineer vs. Data Scientist vs. Software Engineer

Three roles that recruiters and hiring managers confuse constantly. The pay gaps are real and the skill overlaps are smaller than LinkedIn job postings suggest.

*BLS does not track “ML Engineer” as a standalone category. This figure is from Computer and Information Research Scientists (SOC 15-1221), the nearest equivalent.

ML engineers out-earn data scientists by 15-40% at the median, depending on the source. The reason isn’t mystery. Data scientists explore data and build models in notebooks. ML engineers take those models and make them work in production, which means writing deployment pipelines, managing model versioning, handling inference at scale, and debugging performance issues at 2 a.m. when the recommendation system starts returning garbage because a data drift nobody caught shifted the input distribution. The production responsibility commands a premium.

The comparison to software engineers is more nuanced. ML engineers earn a 20-30% premium over general software developers, but the gap narrows at senior levels because a staff software engineer at Google and a staff ML engineer at Google are both pulling $350K+ in total comp. The premium is most visible at the mid-level, where the ML specialization creates a real supply constraint that general software engineering doesn’t have.

For a deeper look at how AI engineering roles compare, including the distinction between ML engineer and AI engineer (which is less clear than job postings pretend), see our AI Engineer Salary Guide.

Skills That Push ML Engineer Pay Higher

Not all ML engineers are paid the same even at the same experience level. The skill set matters, and some specializations carry a measurable wage premium.

PyTorch production experience. This is the single biggest differentiator we see in offer negotiations right now. PyTorch overtook TensorFlow as the framework of choice for research years ago, and it’s been catching up in production deployments. A candidate who can show they’ve deployed PyTorch models to production, not just trained them in a Colab notebook, consistently receives offers $15,000 to $25,000 above candidates with equivalent experience in TensorFlow only. The market has spoken on this one.

LLM fine-tuning and RAG architecture. Eighteen months ago this was a nice-to-have. Now it’s table stakes for senior roles. Retrieval-augmented generation is the pattern behind most enterprise AI deployments in 2026, and candidates who understand vector databases, embedding models, and prompt engineering at an infrastructure level, not just the API wrapper, are commanding premium offers. One of our placements last month, a senior ML engineer moving from a healthcare AI startup to a fintech company, negotiated a $22K base increase specifically because she had built a production RAG system processing 400,000 clinical documents. The fintech company was building something similar for financial compliance and couldn’t find anyone else with that exact experience.

MLOps and deployment infrastructure. Kubernetes, Docker, MLflow, Weights & Biases, SageMaker, Vertex AI. The line between ML engineer and MLOps engineer is blurry, and candidates who sit on both sides of it are the most valuable. If you can build a model AND deploy it AND monitor it in production, you’re doing the work of two roles. Pay reflects that.

Cloud platform depth matters too, but in a less dramatic way. AWS SageMaker fluency vs. GCP Vertex AI vs. Azure ML. Pick one and go deep. Companies rarely switch cloud providers for ML workloads, so they want someone who already knows their stack. A candidate with three years on SageMaker interviewing at a GCP shop will get the offer but might not get the top-of-band number.

Python is universal and non-negotiable. But what separates senior comp from mid-level comp isn’t Python fluency. It’s whether you also know Rust or C++ for performance-critical inference code, CUDA for GPU optimization, or Spark for large-scale data processing. Each one of those is a lever in a compensation discussion.

Industries Paying the Most for ML Talent

Tech companies dominate the top of the pay scale. But they’re not the only game.

Finance and insurance is where ML salaries get surprisingly aggressive. JPMorgan, Goldman Sachs, Two Sigma, Citadel. Quantitative trading firms and large banks have been building ML teams for fraud detection, algorithmic trading, and risk modeling for years, and they pay comp packages that rival big tech. A senior ML engineer at a quant fund can earn $300K-$500K total depending on performance bonuses. The catch is the hours and the culture, which are not for everyone.

Healthcare and life sciences is the growth story. Drug discovery, medical imaging, clinical trial optimization. The ML applications are real and the companies building them, Tempus, Recursion, Flatiron Health, are paying $170K-$230K total for senior ML engineers. Regulatory constraints mean deployment cycles are longer and the work is more careful. Some engineers love that. Others find it suffocating.

Autonomous vehicles and robotics. Waymo, Cruise (what’s left of it), Aurora, plus the robotics companies like Boston Dynamics and Figure. These jobs pay well but the talent pool is extremely specialized. Computer vision and reinforcement learning backgrounds command the highest premiums in this vertical.

E-commerce and retail. Amazon is the obvious one, but Shopify, Instacart, DoorDash, and the recommendation-engine ecosystem around e-commerce collectively employ thousands of ML engineers. The work is less glamorous than autonomous driving. The pay is competitive and the work-life balance is usually better.

Job Market Outlook for ML Engineers

The BLS projects 20% growth for computer and information research scientists through 2034. That’s roughly 3,200 openings per year in the closest equivalent classification. The related data scientist category projects 34% growth with 23,400 annual openings. Both numbers understate the ML-specific demand because BLS occupational categories haven’t caught up to how companies actually hire for this discipline.

LinkedIn’s 2026 Jobs on the Rise report ranks AI Engineer among the fastest-growing roles over the past three years. The World Economic Forum, citing LinkedIn data, reports that AI has already created 1.3 million new roles globally, including ML engineers, forward-deployed engineers, and data annotators, with an additional 600,000 AI-enabled data center jobs.

Here’s the part that matters for hiring: the supply side hasn’t caught up. Graduate programs in machine learning are producing more candidates every year, but the gap between “completed an ML course on Coursera” and “can deploy a model to production and keep it running” is still wide. When we screen ML engineer candidates, probably 60% of resumes that come in for senior roles describe notebook-only experience. The candidate trained a model. Achieved good metrics on a test set. Wrote a paper about it, maybe. Never shipped it. Never dealt with data drift. Never debugged a model that was performing great in staging and returning nonsense in production because the feature engineering pipeline was processing timestamps differently on the production server.

That’s the bottleneck. Not headcount. Production readiness. And it’s why senior ML engineers with deployment experience can negotiate as aggressively as they do.

What Hiring Managers Get Wrong About ML Engineer Compensation

Three mistakes. All preventable.

Posting base salary only. If your total comp package is $180K base + $40K bonus + $30K equity, and you list “$180,000” in the job description, you’re leaving $70K of your selling proposition off the table. Candidates scrolling job boards compare numbers. The company down the street listing “$220K-$250K OTE” is getting the clicks even if the actual packages are equivalent. Always lead with total comp ranges.

Using the wrong title. “Data Scientist” and “ML Engineer” are not interchangeable. If you need someone to build production inference pipelines and you post a data scientist role, the candidates who apply will be great at EDA and Jupyter notebooks and confused when you ask about Docker and CI/CD. Get the title right. If you’re not sure which role you need, talk to a recruiter who specializes in this space before you write the job description.

Benchmarking against the wrong cohort. If you’re a Series B startup in Irvine hiring your first ML engineer, don’t benchmark against Google’s L5 total comp packages. You can’t compete on cash. Compete on scope, ownership, and speed of impact instead. The ML engineer at Google is one of 3,000. The ML engineer at your 40-person company is building the entire ML infrastructure from scratch. Some candidates want that. Price accordingly, but sell the role honestly. We’ve seen more offers accepted at $175K by companies that pitched the scope well than at $210K by companies that couldn’t explain what the ML engineer would actually own.

Things Candidates and Hiring Managers Keep Asking

How much does an entry-level machine learning engineer make?

$120,000 to $147,000 base, roughly, according to Built In and Glassdoor. ZipRecruiter reports a much lower figure around $69K, but that number includes a lot of postings that are really data analyst roles with ML keywords tacked on. A genuine entry-level ML engineer at a company that actually has ML infrastructure, someone with a relevant MS or strong project portfolio, is starting in the $120K range nationally. Higher in California.

Is the ML engineer salary premium over software engineers worth the extra specialization?

Financially, yes. 20-30% more at the mid-level is significant over a career. But that’s not the full picture. ML engineering is a narrower market. Fewer companies hire for it. Fewer roles exist. A software engineer can work at essentially any technology company. An ML engineer needs a company with ML workloads, which narrows the field by roughly 80%. The premium compensates for the reduced optionality. Whether that trade-off makes sense depends on whether you actually like the work or just like the salary number.

Do ML engineers need a PhD?

No. Five years ago, maybe. Most of our ML placements have a master’s degree in computer science, statistics, or a related field. Some have a bachelor’s with strong project portfolios and contributions to open-source ML frameworks. The PhD matters for research-heavy roles at places like DeepMind or OpenAI where you’re publishing papers and pushing the boundaries of what’s possible. For applied ML engineering, building recommendation systems, deploying NLP models, optimizing inference pipelines, production experience beats academic credentials every time. We’ve placed candidates with bachelor’s degrees into $180K+ roles because their GitHub showed real deployment work.

What’s the fastest way to increase ML engineer salary?

$15K-$25K overnight: learn PyTorch production deployment if you only know TensorFlow. The market premium is immediate and verifiable in offer letters.

$20K-$40K over 12 months: build RAG systems or LLM fine-tuning expertise. Every enterprise company is trying to deploy generative AI right now and the supply of engineers who’ve actually done it in production is tiny.

$30K-$60K by switching companies. Internal raises in ML rarely keep pace with market movement. The fastest salary acceleration almost always comes from an external move, especially if you’re at a company that benchmarked your comp two years ago and hasn’t adjusted for how fast this market has moved.

Realistically, how fast can a company fill a senior ML engineer role?

Eight to fourteen weeks for most companies. If you’re in a secondary market, add four weeks. If you require full on-site, add another two to four because you’re eliminating roughly 40% of the senior candidate pool who won’t consider it. We’ve filled them faster, 22 days for a client with a compelling scope and competitive comp, but that’s the exception. The norm is closer to three months. Budget accordingly and start the search before you need the person, not after. We handle these through our AI and ML staffing practice if you want to compress the timeline.

Are ML engineer salaries going to keep rising?

For the next two to three years, almost certainly. BLS projects 20-34% job growth in related categories through 2034. Enterprise AI adoption is accelerating, not plateauing. And the supply of production-ready ML engineers is growing slower than demand. At some point the market will equilibrate. Graduate programs will produce enough candidates, internal training programs will upskill enough software engineers, and AI tools will automate enough of the routine ML work that the supply constraint eases. We’re not there yet. Not close. If you’re hiring, expect to pay at or above the 2026 numbers in this guide for the foreseeable future. If you’re a candidate, your leverage is real, use it, but don’t assume it lasts forever. Markets correct.

For more salary benchmarks across AI and tech roles, try our Salary Benchmark Assistant or browse our IT staffing services to see how KORE1 supports companies building technical teams across Southern California and beyond.