Data Science Careers: Roles, Skills and Entry Paths

Careers · November 2026

“Data science” is used for several jobs that differ a great deal in what they ask of you. A person who builds dashboards for a sales team, a person who trains a prediction model, and a person who designs new algorithms may all carry the title, yet the first needs strong spreadsheet and query skills and the last typically needs a graduate degree. Anyone planning a route into the field gets better results by choosing a specific role first and working backward to the training.

The U.S. Bureau of Labor Statistics (BLS) tracks the clearest of these roles. Its Occupational Outlook Handbook, using May 2025 wage data and 2025–2035 projections, shows strong demand for the core occupations. Projections are estimates, not promises about any one graduate’s job search, but they show where the field is heading.

The Main Roles and What BLS Reports

Data scientists

BLS describes data scientists as people who use analytical tools and methods to draw useful insights from data, which includes cleaning raw data, building machine-learning models and presenting findings to technical and non-technical audiences. The 2025 median pay was $120,230. Employment is projected to grow 35 percent from 2025 to 2035, adding about 95,400 jobs to a base of roughly 275,600. A bachelor’s degree is the typical entry requirement, although BLS notes that some employers want a master’s or doctoral degree. Earlier BLS editions put growth at about 34 percent for 2024–2034, so the direction has held across releases; our retrospective on the data scientist title covers how the job got its reputation.

Operations research analysts

These analysts use mathematical and analytical methods to help organisations solve problems and make decisions, for example in logistics, scheduling and pricing. BLS reports a 2025 median of $88,940 and projected growth of 12 percent over 2025–2035, from about 113,100 jobs. The typical entry-level education is a bachelor’s degree. This is a practical option for students who like mathematics and optimisation more than software engineering.

Computer and information research scientists

This is the research end of the field: inventing and improving computing approaches, including in machine learning. BLS lists a 2025 median of $140,300, projected growth of 22 percent, and a master’s degree as the typical entry requirement, although some federal jobs accept a bachelor’s. It is a smaller occupation, about 38,600 jobs in 2025, so competition for openings is real despite the high growth rate.

Analyst and machine-learning engineer roles

Two common titles do not map neatly onto a single BLS occupation. Data analyst roles focus on querying data, building reports and explaining trends, and are often the first rung for people entering the field. Machine-learning engineer roles focus on putting models into production, which blends data science with software engineering. Because BLS does not publish a dedicated figure for either, treat salary numbers from job sites with caution and check several postings. The AI-specific pay and skills picture is covered in our guide to AI and machine learning careers.

Skills That Recur Across Roles

Choosing Between the Roles

A useful test is to ask what you want your days to look like. If you like finding the story in a messy table and explaining it to a manager, analyst and data scientist roles fit, and the barrier to entry is lower. If you prefer proofs, optimisation and abstract problem-solving, operations research or a research-scientist path suits you better, with more mathematics and, for research roles, more years of study. If you like building systems that run reliably, the machine-learning engineering side rewards software skills over statistics. Job titles overlap between employers, so read the responsibilities in each posting rather than the title, and compare what several employers actually ask for.

Ways In

There is no single route, and the right one depends on where you are starting.

  1. A degree in a quantitative field. Statistics, mathematics, computer science, economics and engineering all lead to these roles. For research-scientist jobs and for some data scientist posts, plan on a master’s degree.
  2. Start as an analyst and grow. Many people reach data science by first doing analytical work in another function, then adding programming and modelling skills.
  3. Bootcamps and certificates. Short programmes can help career changers who already have a degree or work experience. Read placement claims carefully; see our analyses of what placement-rate figures hide and bootcamps versus degrees, and the funding rules in the Workforce Pell guide.
  4. Internships and research placements. A summer role gives you project experience that a résumé otherwise lacks. See when summer 2027 applications open.
  5. A portfolio of finished work. Two or three projects with a clear question, public data, a documented method and a written conclusion show more than a list of tools.

The entry-level market is competitive even where overall growth is high, and a fast percentage does not guarantee openings for beginners. Pick one role, build the skills it lists, and check current postings and the BLS profiles before spending money on training.

← Back to Careers  |  AI and Machine Learning Careers  |  STEM Careers Guide →