Data Engineer vs Data Scientist vs ML Engineer: Titles and Job Search

Data engineer vs data scientist is one of the most common questions from people choosing a direction or trying to search more precisely, and ML engineer usually joins the comparison a moment later. The three roles share tools and vocabulary, so postings blur together. This guide gives a short answer first, then the titles behind each role, how to tell them apart from the description, and how to set up a search that doesn't miss postings because of naming.

The short answer

• Data engineers build and run the systems that move, store and clean data: pipelines, warehouses, streaming, data quality.
• Data scientists use that data to answer questions and make predictions: experiments, statistical analysis, models.
• ML engineers turn models into reliable production software: training pipelines, serving, monitoring.

That's the typical split. Real companies draw the lines in different places, which is why the description matters more than the title.

Data engineer: titles and flavours

Titles you'll see: Data Engineer, Senior Data Engineer, Data Platform Engineer, Big Data Engineer, ETL Developer, Analytics Engineer, "Software Engineer, Data", Data Infrastructure Engineer.

There are roughly three flavours:

• Pipeline and warehouse work: ingesting data from products and third-party tools, modelling it in a warehouse, keeping it fresh and correct.
• Platform work: building the tools other data people use, such as orchestration, streaming, storage and access control. This overlaps heavily with platform engineering.
• Analytics engineering: modelling data for analysts, usually with dbt and a lot of SQL. It sits between data engineering and analysis, and we cover it from the analyst side in data analyst job search.

Interviews usually include SQL, data modelling, a coding exercise in Python or another language, and a design conversation about a pipeline: sources, volumes, failure handling, late data, backfills.

Data scientist: two different jobs under one title

Titles you'll see: Data Scientist, Senior Data Scientist, Applied Scientist, Research Scientist, Decision Scientist, "Data Scientist, Product" or "Data Scientist, Analytics".

In tech companies, "Data Scientist" often means one of two quite different jobs:

• Product or analytics data science: experiment design, causal inference, metrics, forecasting. It's close to senior product analytics and involves little production code.
• Modelling data science: building models for ranking, pricing, fraud or recommendations, often handed to engineers to ship, sometimes shipped by the scientist.

The posting usually gives it away. "A/B tests", "metrics" and "stakeholders" point to the first; "features", "training" and "model performance" to the second. Interviews tend to mix statistics, SQL, a case study and sometimes a take-home or ML theory questions.

ML engineer: where it overlaps

ML Engineer sits between modelling data science and backend or platform engineering. You'll see Machine Learning Engineer, MLOps Engineer, ML Platform Engineer and titles tied to a domain, such as Computer Vision Engineer or NLP Engineer. The AI side of this family, including AI Engineer and LLM roles, has its own guide: AI Engineer, ML Engineer and other AI job titles.

How to tell them apart from a posting

When the title isn't enough, ask:

1. What is the deliverable? A pipeline or table, an analysis or decision, or a running service?
2. Who uses your work? Other engineers and analysts, business leads, or end users through the product?
3. Is there on-call? Data engineers and ML engineers often share responsibility for production systems; data scientists rarely do.
4. What languages and tools are listed first? Spark and Airflow, statistics and experimentation, or containers and model serving?
5. What does success look like? Data freshness and reliability, better decisions, or model quality and latency in production?

Our guide on how to read a job description covers the same habit for any role.

Moving between the three

• From backend engineering: data engineering is often the shortest step. Distributed systems, APIs and databases carry over directly. See the backend developer job search guide for the titles backend engineers usually track.
• From analytics: analytics engineering or product data science, depending on whether you prefer building data models or answering questions with statistics.
• From data science to ML engineering: expect to show production engineering, meaning tests, services, deployment and monitoring, not just notebooks.
• Levels: Senior, Staff and Principal mean different things in different companies, so compare scope rather than prefixes. We explain this in seniority levels in job titles.

Setting up a search that doesn't miss postings

Pick the roles you'd genuinely accept, then track every title that covers them. In Hot Jobs, matching is by keywords in the job title, and it's worth knowing how the data roles split:

• Data Engineer catches Data Engineer, Analytics Engineer, ETL, Data Platform Engineer and Big Data.
• Data Scientist catches Data Scientist, Data Science, Applied Scientist and Research Scientist.
• ML Engineer catches Machine Learning, ML Engineer, MLOps, Deep Learning, Computer Vision and NLP.

Some titles fall elsewhere. "Software Engineer, Data" is matched by Software Engineer (any), and "Data Infrastructure Engineer" by DevOps / Platform / SRE, because that's what the words in the title say. "Decision Scientist" isn't matched by any of these. If you're flexible, pick two or three roles together. Titles with Manager, Director or Head of are left out of all three, and junior titles are hidden unless you turn them on.

Hot Jobs checks 390+ company career pages every 30 minutes, plus the Remotive and Himalayas boards, and sends new postings for the data roles you pick to Telegram. €3 once, no subscription. Connect the bot