Joanna Kochel works at OTA Insight as a Data Science Engineer. After completing her undergraduate with an Engineer's degree at Wroclaw University of Science and Technology, she started her career at Nokia in Poland. Since then, she has worked internationally in Paris and at her current role with OTA Insight in Belgium.
With more than six years of experience as both a programmer and data scientist, Joanna is an inspiration for young and aspiring data scientists. We recently interviewed her to learn more about her journey, her career history, and what the Young ICT Lady of the Year nomination means to her.
It's actually a funny story. So to answer the first question - yes, I always wanted to do mathematics. Did I think from the beginning to study mathematics as such? Not really. I come from a small town in a rather traditional part of Poland. As a child I was always good at calculus and was lucky to have amazing teachers, who happened to be (up till high school) all women.
I think this fact was a key factor to show me that mathematics is also for girls. But the stereotypes were quite deep in my head, so when I was a child I wanted to be… an accountant. When I think about it now it seems silly. After all, who as a child, wants to be such a "down to earth" professional, but the only math-related profession (not being teaching - since I knew I would not like to do that) done by women at that time was accounting.
While growing up I enjoyed studying math more and more. When I was 15 and had to choose a high school and was advised by a friend of mine to enroll at one of the best in Poland, that was focused on mathematics. I passed the tests and moved 300km from home to a boarding school - still having accountancy in my head. And that was the breaking point.
It was a big city and really smart kids from all around the country. And their dreams were also bigger than where I came from, which made me wonder about my future. I received 12 hours of mathematics schooling a week and additional 4 hours of programming. The teachers showed me there is much more to mathematics than statistics and calculus.
What I fell in love with is how math is the foundation of programming and how you can apply it - through technology - in almost every aspect of life. This made it and easy decision on my part in terms of what to study.
To be honest, for quite some time I didn't know myself. My first job was actually related to pure programming and then, because of my study field, I moved to Data Analytics. But because of my coding background, I don't think I ever worked as a standard Data Analyst, but more something in between data engineering and data analytics. What does it mean? That I had two parts to my job.
The first one being programmatically cleaning and organising the data in the databases - which I would call the data engineering part, and the second part was "pure" data analytics. The latter is, depending on the company, working with internal or external clients and basically providing them insights based on the data.
In my vision in comes down to a few steps:
So as you can see there is a lot of communication and the technical part is actually not the most important factor.
Data Science is something more focused on innovation and requires bigger technical skills. Data Scientists, at least in the companies I have worked for, are mostly focused on the research part. They try to work on less defined and more complex problems.
Often there are no clear tasks and no strict deadlines. There is always a big chance that most of your work will be trashed, because your idea didn't work. But when things work, the feeling of accomplishment is really big.
You also need quite a lot of domain knowledge (currently for me this is in the hospitality industry), and you need to know quite some math to be able to validate your ideas in a statistically meaningful way and you also need to know how to code or use ML algorithms. So it's a much more independent and technically demanding job than DA.
So if you ask me to answer this question briefly:
As a Data Analyst, a big part of your job is to gather requirements and then transform (rather structured) data into digestible insights.
As a Data Scientist you get a problem statement and a bunch of (usually messy) data sources and you need to work it out yourself with bits of feedback from different directions.
Fortunately, most of the skills I use at work I learned at the university. When it comes to hard skills - I had many courses on programming, statistics, probability, logic, time series analysis, algorithms. I think the only missing part was Machine Learning related subjects, which I had to learn on the go.
The good news is that on this topic there are endless online resources, so I just tried to get some high level overview early on and then learned things while working.
I was also lucky enough to gather experience in communication, time management and projects during my studies. As a president of my faculty's Pure and Applied mathematics, I had to work with different people: from students in other organisations and external companies to the university authorities. It taught me a lot about working with various stakeholders and is still very useful at work these days.
My "nuclear" team in the Data Science team. It's the place where we share our ideas and brainstorm around them a lot. The guys I work with are incredibly smart, so it's always intellectually challenging - in a good way of course! But we also work a lot across different teams.
On a daily basis. The Data Science team is an inseparable part of the broader engineering team at OTA. We take part in the standard planning and we participate in the product developments if it's data-related. We also work a lot with other teams. I personally work mostly with two of them.
One being the Product Team, which I’ve worked extensively with over the past 1.5 years to build our new tool, Market Insight. I worked on the data side of this product and also on ways we to ingest the data and make it understandable and actionable.
This was one of the biggest challenges in my career, for two reasons. One being the fact that there was a lot of communication on different levels involved: product manager, design team, engineering, data providers, front-end and back-end engineers and to not get lost in all of it was really something.
The second thing was that it was a technically challenging project: we had to explore many different paths and many of them were dead ends. We had to learn a lot of new technical frameworks and read research papers to find the best solutions. We had to get far outside of our comfort zone.
But all the effort fortunately led to a great final product, which is helping hoteliers on their way to recovery from the pandemic.
The second team I work with is the Marketing Team. This year we've put extensive work into generating data driven marketing content. Every time I see something interesting in the data which could be potentially used for the broader public, we catch up with the team and try to figure out how we can present it best.
This is how we started a data project based on hotels openings rates and the series of blog posts connected to Covid-related search for flights and hotels behaviours.
We start every day with a stand up among the Data Science team. We talk about the work from the previous day - usually showing some results and getting feedback or brainstorming when we are stuck on something. We also discuss the plan for the day and then we go each in that direction, since we often work on different projects with different teams.
After that I usually have a meeting or two a day to discuss progress on ongoing research topics. Besides that most of my day is working with code (mostly Python) to implement and validate some ideas. I also leave a bit of time for writing down the results.
The first piece of advice is to try not to get blocked by stereotypes. That's why I find the idea of this award so important. There are still not many women in STEM fields. There is a lot of research being done on what is causing this.
But what I think is a very important factor to improve the ratio. As mentioned it was incredibly important for me to have role models. It's important to see that there are women in STEM and they are doing amazing things. It's a great eye-opener and motivator.
Another bit of advice is to always take chances. I took some risks in my life (like moving to Paris without speaking French), but all of these experiences, even if not always successful, have taught me a lot.
On the technical side, I think Data Science is not an easy field to enter without a university background in math-related fields. It’s not mandatory though, as there are many resources available online now.
There are also vibrant communities of women in tech-related fields where you can find support. One example is Geek Girls Carrots , where I have volunteered, and in Belgium women.code(be). Having someone to relate to and connect with always makes things easier!
Considering that I moved to Belgium just over a year ago, I didn’t anticipate this. When I saw my name alongside the other nominees, I was honored to be included among the finalists.
I also find the award itself important and I am happy that something like this exists in Belgium. As mentioned before, I think role models are extremely useful.
I believe that seeing other women “up there” will tip the balance for many younger girls to go in the direction of IT.
The election for Young ICT Lady of the Year awards runs from December 14th until December 31st.
Cast your vote for Joanna here.