From physicist to Data Scientist — what actually made the difference.
I recently pulled out the CV I wrote six years ago — the one that got me my first data science job at the BBC. Reading it back, I was struck by what it didn't have: publications, conference wins, a computer science background. What it did have was a clear narrative, honest momentum, and evidence of someone who could figure things out.
A qualification alone — regardless of quality — isn't enough. For people coming from theoretical fields like physics, maths, or chemistry, organisations want evidence of how you've applied that knowledge. Professional contexts differ from academic ones, and companies care about talent that works for them.
Initiative is critical. You must demonstrate the ability to translate theoretical understanding into practical value — through projects, placements, or self-directed learning. That's what establishes you as someone capable of applying knowledge toward real organisational challenges.
Leaving a doctoral programme midway was difficult. But I recognised I preferred applying skills to building functional solutions that generated tangible outcomes, rather than publishable contributions. So I leaned into what I had:
My professional summary showed clarity and assurance — not overclaiming, but displaying determination, motivation, and the ability to work independently.
My doctoral research in computational chemistry wasn't technically machine learning — but it was deeply technical and investigative. I was programming, processing large datasets, configuring experiments, and communicating results. I reformulated this using data science terminology, making the research immediately understandable to recruiters unfamiliar with my field.
The reframing turned scholarly work into measurable, data-oriented accomplishments.
I included everything pertinent from my education — modules, applications, and small coding projects. Anything demonstrating quantitative reasoning or data management qualified. From my postgraduate work, I featured principal component analysis, stochastic methods, and a Java-based modelling project.
If you're early in your career, don't dismiss academic projects as "not real experience." They demonstrate you can think and build.
Beyond pure technical capability, I showed I could communicate, collaborate, and deliver. Throughout the CV:
I was honest about where I was still learning. Where I was still acquiring statistical methods, I wrote "Familiar with supervised learning algorithms and exploratory data analysis." Honesty about development, paired with visible enthusiasm, signals the right kind of trajectory.
One to two pages. Ordered presentation. Purposeful sequence. Visual clarity. Recruiters examine hundreds of applications and many use keyword searches to filter. Organisation isn't decorative — it communicates that you think clearly and respect the reader's time.
My placement at Procter & Gamble supplied practical legitimacy. Instead of listing responsibilities, I showed impact and independence. That section verified I could own projects, deliver outcomes, and work at pace — all essential for analytical roles.
Rather than submitting and waiting, I found the company's Data Science blog, researched what they were working on, and then personally contacted a recruiter — with a sincere message and my materials. I communicated the skills I'd gained, what inspired me about their work, and how I could contribute.
That focused initiative — reaching out, showing awareness, and making a genuine human connection — is what distinguished my application.
Looking back, this CV worked because it showed:
I didn't claim comprehensive expertise. I showed I was learning fast, taking initiative, and celebrating small wins along the way. That substance — momentum — is what hiring managers care about. Not perfection.
Your CV isn't a list of jobs. It's the story of your intellectual development, your capability growth, and where you're headed.
You don't need flawless qualifications. You need focus, honesty, and willingness to learn fast.