mlengineersalary.com
section 7 : role comparison

In [11]: # mle_vs_ds.ipynb

ML Engineer vs Data Scientistwhy MLE base is 15-40 percent higher

7.1The headline gap

fig. 7.1

ML engineer

$173,000

median base salary

+$35,000

MLE premium

+25.4% gap

Data scientist

$138,000

median base salary

Figure 7.1. Market-comp self-report, not BLS. Synthesised from public self-report aggregators (Levels.fyi, Blind and comparable disclosures), captured May 2026. Self-reported samples skew toward higher-paid respondents. Note that "data scientist" here is the market-comp job title, not the SOC. Section 7.1a gives the official wage series.

7.1a The same comparison in official wage data

BLS OEWS May 2025

BLS publishes no machine learning engineer occupation, so it cannot price this gap directly. What it does publish is the data scientist series itself, SOC 15-2051 Data Scientists, at a national median of $120,230 across 262,440 jobs, and the occupation that absorbs most production ML engineering, SOC 15-1252 Software Developers, at $135,980 across 1,687,890 jobs. That gap, roughly 13.1 percent, is the closest official proxy for the premium above, and it is narrower than the market-comp figure.

BLS Occupational Employment and Wage Statistics, May 2025, SOC 15-2051. Snapshot taken 6 September 2026. BLS Occupational Employment and Wage Statistics, May 2025, SOC 15-1252. Snapshot taken 6 September 2026. Public domain; attribute to the Bureau of Labor Statistics.

7.2Side-by-side comparison

8 metrics
MetricML engineerData scientist
Median base salary$173,000$138,000
Median total compensation$245,000$180,000
Entry-level base$100k - $140k$85k - $115k
Senior level base$180k - $230k$155k - $185k
Senior TC at T2 hyperscaler$280k - $450k$200k - $360k
Open US roles (2026)~45,000~62,000
Required degreeCS / Eng preferredStats / Math / CS
Avg. years to senior6 - 8 years5 - 7 years

Table 7.2. Market-comp self-report, not BLS. Synthesised from public self-report aggregators (Levels.fyi, Blind and comparable disclosures), captured May 2026. Self-reported samples skew toward higher-paid respondents. Role counts and time-to-senior are editorial estimates from job-board observation, not survey statistics. No row in this table is a BLS figure.

7.3Skills required

9 skills
SkillMLEDS
Software engineering·
Production ML systems·
Distributed computing·
Model optimisation
Statistical analysis
Data visualisation·
Business communication·
SQL / data querying
Experiment design·

7.4Day-to-day responsibilities

ML engineer

  • Design and train production ML models
  • Build training and serving infrastructure
  • Optimise inference latency and throughput
  • Write production-grade code (Python, C++)
  • Collaborate with platform and software teams
  • Own models from experiment to deployment
  • Performance-engineer training loops
  • Define ML system architecture

Data scientist

  • Analyse large datasets for business insights
  • Build statistical models and experiments
  • Create dashboards and visualisations
  • Run A/B tests on product features
  • Communicate findings to stakeholders
  • Generate business recommendations
  • Exploratory data analysis
  • Partner with product and business teams

7.5Frequently asked

3 questions

Q.Why do ML engineers earn more than data scientists?

A.ML engineers command higher salaries because they require strong software engineering skills on top of ML knowledge. They own the full pipeline from training to production deployment, requiring systems thinking, performance optimisation, and reliability expertise that data scientists typically don't need.

Q.Should I become an ML engineer or data scientist?

A.If you enjoy coding, system design, and shipping production systems, MLE is the higher-paying path. If you prefer analysis, statistics, and business storytelling, data science suits you better. Many DS practitioners transition into MLE roles after building engineering skills, often picking up an immediate 15-25 percent salary lift.

Q.Is the data science to ML engineer transition common?

A.Very common. Many ML engineers started as data scientists and upskilled in software engineering, distributed systems, and MLOps. The transition typically takes 1 to 2 years of deliberate practice and comes with an immediate 15 to 25 percent salary bump on the role change.

→ MLE salary by level→ ML specialisations

In [ ]: # reader_questions.ipynb

Ask about a band this preprint does not cover

Write in. A person opens every message, and the ones we can answer come back with the survey release or the public disclosure the number is drawn from, stated plainly enough that you can go and check it. We put no clock on a reply. Should an answer be worth reading twice we might set it out here stripped of anything identifying, and we will write and ask you before that happens.

Out[ ]:

Treat everything on this site as compensation reference assembled from published surveys and public filings, not as financial advice and not as a benchmark any single negotiation should be anchored to. Your own offer turns on the employer, the level it is pitched at, the equity instrument behind it and the market on the day it is made.

Write to the deskquestions@mlengineersalary.com