Start with the work, not the label
Choose one question about the daily work: whether you enjoy experimentation, building systems, explaining evidence or coordinating decisions. Job titles are too broad to answer that for you.
Design a low-cost test
Use a project, society role, short placement, informational conversation or public dataset to recreate one part of the work. Define what you want to notice before you begin.
A career experiment should reduce uncertainty, not prove a lifelong identity.
Record what changed
Write down what gave you energy, what felt frustrating and which skills you want to develop next. Use that evidence to choose the next experiment.
Compare the reality with your assumptions
Before the experiment, write down what you expect to enjoy, what you think will be difficult and what a normal week might contain. Afterwards, compare those assumptions with what you observed. A useful test can make a direction more attractive, less attractive or simply more specific.
Talk to someone doing the work and ask about recent tasks rather than a general job description. Questions about decisions, collaborators, deadlines and trade-offs reveal far more than asking whether they enjoy their career.
Turn the result into a next step
If the work still interests you, increase the realism gradually: move from reading to a short project, from a project to shadowing, and from shadowing to a placement or research experience. If it does not, name the part that failed the test and carry the useful evidence into your next exploration.
The goal is not certainty. It is a better-informed decision and a clearer idea of what to test next.
A good career experiment produces evidence, even when the answer is “not this.”