Associate Professor Marian-Andrei Rizoiu shares what data scientists do, where they work and what drives discovery in the field.

Every day, vast amounts of data are created across finance, healthcare, technology and science. On its own, that data is messy, fragmented and often unusable. Making sense of it and turning it into insight is the work of a data scientist.

Associate Professor Marian-Andrei Rizoiu leads the Behavioural Data Science Lab at the University of Technology Sydney (UTS). He describes data science as a discipline grounded in scientific thinking, where scale and complexity are part of the challenge.

“A data scientist is essentially a scientist who uses data,” he says. “Our job is to work through very large volumes of information and uncover what’s meaningful within it.”

He compares the process to mining. If data is the ore taken from the ground, information is the gold within it. Data scientists are the ones who extract that gold. 

What is data science and what do data scientists do? 

Data science focuses on extracting information from large datasets and using it to explain patterns or make predictions. It is used wherever large volumes of data are generated. 

In finance, data scientists help keep people safe from scams. In medicine, they analyse genetic and medical data to understand which parts of DNA and chromosomes are linked to particular conditions. Increasingly, data science is also used to understand human behaviour at scale, from how information spreads online to how people respond to health or policy interventions.

Data scientists’ work can include:

  • Collecting large datasets
  • Cleaning data and fixing inconsistencies
  • Conducting descriptive analysis to explore distributions and patterns
  • Breaking data into groups to understand how it behaves
  • Using machine learning and artificial intelligence to explain data or predict future outcomes

Much of a data scientist’s time is spent preparing data before analysis can begin. Real-world data often contains errors caused by equipment malfunctions or misalignment. These issues must be corrected before the data is usable.

Once the data is clean, analysis begins. This usually starts with descriptive analysis and progresses to predictive analytics, where models are used to explain what has been observed or anticipate what might happen next.

A typical day might involve cleaning data, testing models, discussing results with collaborators and refining questions based on what the data reveals.

Getting started in data science 

Becoming a data scientist typically requires a university degree. This usually involves three to four years of undergraduate study, sometimes followed by a master’s degree. For those working on cutting-edge scientific problems, a PhD is often required. 

Training in data science draws on several disciplines, including: 

  • Mathematics and statistics 
  • Computer science 
  • Programming and database systems 
  • Scientific and analytical languages 

Rizoiu notes that advanced research roles, particularly those tackling unresolved scientific questions, generally require doctoral study. He also emphasises the importance of curiosity, patience and persistence, as progress often comes through trial, error and refinement.

Where data scientists work

The demand for data scientists is high, largely because data in digital form is everywhere. Some data scientists, particularly those with PhDs, work in universities, where they combine teaching with research. 

Most people trained in data science work in industry. In these roles, they may be embedded within organisations or lead data science teams. They are often involved in developing new products, extracting insights from data and creating data-driven tools and applications that are passed on to consumers.

Data scientists typically work alongside others, applying their skills in collaboration with researchers, engineers or domain experts. Rizoiu describes data science as a versatile set of skills, with opportunities to work in both research-focused and applied settings.

For Rizoiu, the most compelling part of data science is discovery.

“There’s a unique excitement in realising you’re the first person to know the answer to a question,” he says. “Even if it only lasts for a few minutes, that moment is incredibly powerful.”

It’s that sense of curiosity, and the drive to uncover something previously hidden that sustains the work, even when the process is complex or demanding.

Thinking about a future in data science? 

Data science is a field with growing demand and wide-ranging applications. For people who enjoy problem solving, working with technology and exploring unanswered questions, it offers a career that can evolve across industries and roles.

As data continues to shape how decisions are made and systems are designed, the ability to understand and interpret it will only become more valuable.

Share

Contributor

Portrait of Dr Marian-Andrei Rizoiu

Dr Marian-Andrei Rizoiu

Associate Professor

The Data Science Institute

how to become a data scientist thumbnail

Associate Professor Marian-Andrei Rizoiu on becoming a data scientist

From what data scientists actually do, to how to get started, where the role can take you and what makes the work so rewarding.

Associate Professor Marian-Andrei Rizoiu on becoming a data scientist transcript

What it’s really like to be a data scientist 
with 
Associate Professor Marian-Andrei Rizoiu 
 
Duration 4min 42sec 
 
00:00:00:02 - 00:00:14:06 
 
Hello. My name is Marian-Andrei Rizoiu and I lead the behavioral Data Science Lab at UTS. I'm here to answer your questions about how to become and what does it mean to be a data scientist in, in a day to day life. So let's kick it off. 
 
00:00:14:06 - 00:00:16:07 
 
What is a data scientist? 
 
00:00:16:07 - 00:00:20:10 
 
Well, a data scientist is a scientist that uses data. Essentially 
 
00:00:20:10 - 00:00:32:07 
 
we are scientists that can sift through very, very large volumes of data. And within that data there is the information. To use an analogy, if the data is the 
 
00:00:32:07 - 00:00:42:06 
 
ore that you dig out of the ground and the information is the gold that you want to extract, we are the people that extract the gold from the very large volumes of data. 
 
00:00:42:06 - 00:00:52:15 
 
Where are we useful? We are useful, wherever large volumes of data are being generated. In the financial world, to keep you, keep you safe from scams. 
 
00:00:52:15 - 00:01:02:14 
 
In medicine, to understand which parts of our DNA and chromosomes are responsible for certain, for certain afflictions and so on and so forth. 
 
00:01:02:14 - 00:01:04:13 
 
How can I become a data scientist? 
 
00:01:04:13 - 00:01:26:14 
 
You typically require a university degree, a bachelors of three, four years of, sometimes a master's, many times also a PhD. So anywhere in between 3 to 8 years of, of study where you look at things around, typical science or how to do statistics, how to, you need mathematics, but also a lot of computer science and computer science skills. 
 
00:01:26:14 - 00:01:34:12 
 
So that would be programming databases, and scientific languages. Being able to use scientific languages. 
 
00:01:34:12 - 00:01:43:13 
 
For the scientists, which are really on the cutting edge of, of technology that we solve those unsolvable questions typically you also require a PhD. 
 
00:01:43:13 - 00:01:46:07 
 
That's how you become a data scientist. 
 
00:01:46:07 - 00:01:51:06 
 
What does a data scientist do in their day-to-day work? 
 
00:01:51:06 - 00:02:05:16 
 
A lot of time is actually spent on collecting the data, cleaning the data, making sure that it's consistent and correct. You won't believe the amount of inconsistencies that that, that get into the data while we are collecting it. 
 
00:02:05:18 - 00:02:13:02 
 
Equipment malfunctions, they are misaligned, just to name a couple of, of these. So all of these errors, they need to be cleaned 
 
00:02:13:02 - 00:02:23:12 
 
before the data is usable. Once the data is clean, typically a data scientist will then move to what we call a descriptive analysis. So we start looking at distributions. 
 
00:02:23:12 - 00:02:49:03 
 
We start breaking it into groups. Trying to make sense out of that. Once we have finished at this step we move into predictive analytics. Predictive analytics typically use machine learning and artificial intelligence tools to be able to either explain what we have collected, the data we have collected, or make predictions about the future. As I was giving an example earlier, if I'm collecting a lot of data about a human genome, 
 
00:02:49:03 - 00:03:00:01 
 
and I'm studying a particular medical condition, can I link the the combinations of genes to that particular, affliction so that I can easily flag it in new patients. 
 
00:03:00:01 - 00:03:03:02 
 
What's your favourite part about being a data scientist? 
 
00:03:03:02 - 00:03:25:19 
 
I personally have always been excited about finding new things. Being able to know something new and knowing that I am the first person to know it. I really love the excitement of holding for that precious couple of minutes. The knowledge that you are the first person in the world to know the answer to a particular question, and that's what got me thinking. 
 
00:03:25:19 - 00:03:28:14 
 
What different settings can data scientists work in? 
 
00:03:28:14 - 00:03:40:06 
 
The demand for data scientists is actually very large because data and data in digital format is all around us. So yes, some of the data scientists, particularly the ones with PhDs, will be doing academic research. 
 
00:03:40:06 - 00:04:06:11 
 
So be, do teaching and research within an academic environment. But they are a minority. Actually, most of the people trained in data science, they will work within the industry, within the corporate environment where they will be embedded or leading data science teams. They are the people that will be developing the new products. They are the people that build the innovations that that then are passed on to consumers. 
 
00:04:06:13 - 00:04:17:04 
Unknown 
So the data scientist is actually a very versatile, set of skills to have because you can be doing cutting edge research and within the, 
 
00:04:17:04 - 00:04:26:08 
 
scientific approach. But you can also work very applied to develop new products, to develop new insights, new banking applications. There's a lot of opportunities. 
 
00:04:26:08 - 00:04:39:22 
 
So if I have convinced you that the data scientist career is the right for you, just hop on to the UTS website to check what opportunities and courses are there available for you to become a data scientist.

Interested in Data Science?

Explore UTS courses in Data Science to find out where this field can take you.

Student stories and news

Webpage

Marni is one of just a handful of Indigenous architecture graduates working in Australia today.

Webpage

Learn how Rhiannon Brownbill uses architecture to create more humane and healing healthcare spaces.

Webpage

Safwan is proof that your first career doesn’t need to be your only career.

Webpage

UTS’s Professor Chris Turney shares his thoughts on the careers and job titles that have the most potential to act on climate change now and into the future.