Career Pivot
Data and AI are crowded with promises. This session sticks to the work.
Analysts, engineers, finance and operations people, marketers and others are looking at data and AI roles because the field is in the news. This session explains what the roles involve, how they differ, what a realistic move from another field looks like, and how to tell a useful route from an expensive one.
Explore this page6 sections
Strong demand does not mean an easy entry for every career changer.
The World Economic Forum's Future of Jobs Report 2025 lists big data specialists and AI and machine learning specialists among the fastest-growing roles to 2030, in percentage terms. That is a forecast of growth from a fairly small base, not a count of vacancies today, and it says nothing about how easy it is for any one person to be hired into them.
Hiring for these roles varies a great deal between employers. Some want advanced degrees and research ability, some want practical analysis and engineering, and some want people who understand a business problem and can use data tools well. Which one is meant is rarely clear from a job title.
For people coming from another field, the most realistic move is often into the adjacent role that combines their existing knowledge with data skills, rather than straight into the most technical one.
A grounded way into data and AI
35 topics across 7 areas. Seven parts, from understanding the roles to checking what is promised.
The Different RolesThe names overlap, but the work differs.5
- Data analyst: questions answered with data, reports, dashboards and decisions
- Data engineer: pipelines and systems that move and store data reliably
- Data scientist: modelling and statistics, with business problems attached
- Machine learning engineer: putting models into working products
- Roles that use AI tools in another function, such as marketing or finance
Read several job descriptions for each. Titles differ between companies, and the same title can mean different work.
Which Roles Are Closer to Where You AreStart from what you already know.5
Somebody with years in healthcare operations who learns analysis brings knowledge of the domain that a team in that field may value. That is worth testing by talking to such teams, and it is not a guarantee of a role.
- Analysis roles inside your own function, such as finance, operations or marketing analytics
- Data-adjacent roles where domain knowledge matters more
- Technical roles, if you have software or maths experience
- Working with data in your current job first, where possible
- Where your existing field's knowledge is the advantage
The FoundationsWhat is usually needed, whatever the role.5
- Spreadsheets and clear thinking about numbers
- SQL for working with data
- A programming language, commonly Python, for most analysis and modelling roles
- Basic statistics and probability
- Communicating findings to people who are not technical
Evidence Beyond a CertificateWhat hiring managers can actually see.5
- One or two real projects on real, messy data
- A write-up that explains the question, method and limits
- Work you have done with data in your current job
- Code or analysis other people can read and run
- Being able to explain every step in an interview
A project on the most common practice datasets is the same as everyone else's. A question from your own field is more convincing.
Using AI Tools Versus Building ThemTwo very different things.5
- Using AI tools well in your own work is a skill in almost every field
- Building and training models needs mathematics and engineering depth
- Where prompts, automation and evaluation sit in between
- Being clear which of these a job actually requires
- Not claiming more than you have done
Courses and ProgrammesUseful for some things, not a ticket.5
- What a good course can give: structure, practice and feedback
- What it cannot give: experience, judgement, a job
- Free and low-cost options worth trying first
- Warning signs of over-promising, covered in a separate session
- Checking what recent participants actually went on to do, from sources you can verify
A Realistic PlanMonths, not weeks.5
- A first project within the first month
- Foundations first, then projects that use them
- Talking to two or three people in the target role
- Applying for adjacent roles while continuing to learn
- A review point at three and six months
Common mistakes
- Paying a large amount for a course before testing the interest
- Collecting certificates instead of building projects
- Aiming only at the most technical role
- Copying the same practice projects as everybody else
- Believing guaranteed job claims
- Leaving a stable job before checking the entry route
What participants leave with
- A map of roles with the nearest ones marked
- A starter learning plan with a first project
- A checklist for judging courses
- A way to describe their domain knowledge as an advantage
- Questions for people already in the field
How the session runs
A practitioner who works with data or AI, ideally one who came from another field, explains the roles and what a hiring manager looks at. Participants then look at real job descriptions together and map their own experience against them. Evening or weekend sessions, in person or online. Speakers are asked not to promote a course or a company.
What your students leave with
- A clear map of the different kinds of data and AI roles
- A view on which roles are closer to the work you already do
- Knowing what a hiring manager looks for beyond a certificate
- A way to build evidence with one real project
- The ability to spot an over-promised course or programme
- A step-by-step plan that does not depend on a guaranteed outcome
Scheduled sessions
Nothing scheduled yet
Sessions are arranged with a college once a date is agreed. Ask us and we will find the right person for it.
A student rather than a college? See what is coming up, or ask your placement team to host this.
Start from the work, not from the headlines.
Tell us who your students are and what stage they are at. Sessions are free for participants.