Inspirational Woman: Karin Sasaki | Senior Consultant, Ekimetrics

Karin SasakiKarin is a mathematician with a Ph.D. and five years’ experience in modelling and data analysis in various industry and academic settings.

She has worked with a variety of data that has come from molecular biology systems, as well as from operational research, and now marketing. Her specific modelling and analytical skills include low dimensional topology, topological data analysis and machine learning.

Tell us a bit about yourself, background and your current role

I’ve moved around a lot. To give you a very quick history, I was born in Colombia and came to the UK with my family when I was 13 years old. I stayed in England through school and then did my undergraduate and postgraduate studies in mathematics at Imperial College London. From there I moved to Germany, where I worked as a consultant in modelling techniques for academic research in various areas of molecular biology. Between 2017 and 2021, I shifted my career focus from academia into data science, and ultimately landed in the data industry, where you find me today.

I’m currently a Senior Consultant at Ekimetrics. My most recent work includes doing an in-depth investigation on computer recognition: I’m looking at how a computer can be trained to accurately describe images and videos. Using the results, we’ll then be able to answer questions we have around marketing effectiveness, understanding what type of visual content works best across different social platforms.

Did you ever sit down and plan your career?

No. I have changed my career direction twice (from pure to applied mathematics initially, and then from academic research to working in the data industry). Both times I was simply following my interests. And in both cases, I was inspired by the potential of data and the opportunity to use and develop exciting methodologies in data science.

Have you faced any career challenges along the way and how did you overcome these?

Data processing and analysis is a large part of every data scientist’s job, but I tend to feel unfulfilled when that’s all I do. The most fulfilling jobs for me are those that require self-learning and development as part of the job.

It can also be tough when the goals of a data science team aren’t on the radar of senior management. If leaders don’t provide data scientists with the information required about the data they’re using, scientists resort to filling in the gaps themselves, often having to make uncomfortable assumptions.

In the worst-case scenario, the data science team spends time and effort developing methodologies that are not used or deemed to be useful by the team of people for whom it was originally intended. This can also lead to the data team being judged and their true value isn’t realised. It’s why support and understanding from senior leaders are so vital.

In terms of overcoming challenges, I always try to find a solution to accommodate my needs and communicate with managers if there’s a problem. There are many ways to go about it. You can look at how to do things differently within your team, perhaps trying a different distribution of work so that everyone has more variety.

It’s also necessary that team managers know the interests and strengths of their team members. Not just in terms of skills but also their aspirations. For example, does this person eventually aspire to become a project manager, or do they work best in the more technical aspects of projects? And how can we value both types of people, and use their strengths?

Alignment is also key. Since a data science team is usually (if not always) part of an ecosystem, it is important that its internal goals are aligned with those of the more global teams.

This can translate to any other data science teams across the company, too. Connecting with them, seeing how you could possibly share knowledge and even collaborate cuts problems tenfold. Beyond that, secondments or even permanent placements can be a really useful way of sharing skills and problem-solving.

What has been your biggest career achievement to date?

I feel happy that I’ve been given the opportunity to work at Ekimetrics, because the company uses data science in ways that I always hoped I would. I feel proud to have developed myself and to have worked my way up.

What one thing do you believe has been a major factor in you achieving success?

Having the support to be able to follow my interests is the big one. This is partly in my home environment with family members who contribute to the development and care of my son. And also at work: colleagues give me their trust and support, even in situations where I may not have directly relevant experience or I may need to work in patterns that are different to a normal 9-5 routine.

What top tips would you give to an individual who is trying to excel in their career in technology?

The best way I’ve developed is on the job. I’d advise someone who’s interested in pursuing a career in data science to find somewhere where their development is supported and there’s time specifically allocated for learning. It’s much better than studying alone.

You will ultimately gain more experience with real and practical applications of the theory. Plus, you’ll be working alongside people with more experience, who you can watch and learn from. Working as part of a team, rather than in isolation, also helps you to do work faster and to a higher standard.

Do you believe there are still barriers for success for women working in tech, if so, how can these barriers be overcome?

Perhaps. Fortunately, I don’t think I’ve personally experienced being at a disadvantage in the workplace because of my gender or race. I’m also currently working in a place where there’s a 50:50 split in gender, which is a truly impressive achievement, given that the proportion of women in STEM subjects was around 24% in 2019.

I do think it’s difficult to be and become truly conscious of the various biases we may have, and how those may affect the decisions we make – including when it comes to hiring, promoting, or collaborating with others. In those cases, where decisions affect another person’s development or career progression, I would strive to adopt a strategy that both finds the best result for the company as well as positively discriminates towards those underrepresented groups in the company.

For example, in hiring, I would prepare a concrete list of things to test, and then create a certain acceptance threshold. For the pool of candidates that are above that threshold, I would positively discriminate towards underrepresented groups in my company or team – rather than purely hiring based on “quality” and offering the position to the very “top” candidate. This contributes to stronger teams that can come at problems from different angles to solve them more easily.

What do you think companies can do to support and progress the careers of women working in technology?

It’s not always easy to overcome internal bias, but it’s vital we don’t let those biases define our actions and decisions. Making the conscious decisions to prioritise equal gender representation in companies, and across all seniority levels, is the first step.

Having well-thought-out mentorship schemes, where younger people in the company can shadow/learn/work with from more senior members, is also a great step forward. It sets a standard for what’s expected and creates a culture of equality.

More relevant to women who already have families – as well as men who now take up more of the childcare/home duties – is flexible working. It’s not always possible parents to be available solidly between 9am and 6:30pm every day. If they keep up with the work, I see no reason why work schedules cannot be a little flexible. And I know this doesn’t only apply to parents: I think many people can benefit from finding the right working style that suits them.

There are currently only 21 per cent of women working in tech, if you could wave a magic wand, what is the one thing you would do to accelerate the pace of change for women in the industry?

I’d change how both candidates and companies think about that challenge – and that it seems unsurmountable.

Regarding the candidate viewpoint – I feel that many young women can be put off a career in tech because they don’t believe the career is open to them. They may be put off by the stereotype image of a person in tech: a man who works in isolation most of the time. Or be worried about the culture, environment, or lack of opportunities.

In reality, tech businesses desperately want to attract more women to challenge those perceptions and to create more diverse businesses. The opportunities are out there.

It’s also important to say that the idea that women are any less reliable in the long term because they may choose to have a family is wildly inaccurate. In fact, parents (both men and women) by necessity need to become more efficient in the blink of an eye. This means not only that they learn to juggle home and work, but they work more efficiently (as many working parents will tell you!). It’s a misconception that we need to deal with to help redress the gender balance in the workplace.

What resources do you recommend for women working in tech?

Podcasts:

  • Freakonomics (freakonomics.com/podcasts/)
  • Data Skeptic (dataskeptic.com)
  • How to Think Like and Economist Podcast
  • Sawbones (not really career relevant, just interesting maximumfun.org/podcasts/sawbones/)

Books I’ve read recently, (or I’m currently reading):  

  • Computer Age Statistical Inference (by Bradley Efron and Trevor Hastie)
  • Doing Data Science (by Cathy O’Neil and Rachel Schutt)
  • How to Win Friends and Influence People (by Dale Carnegie) – The title sounds more manipulative than the actual content!
  • Thinking, Fast and Slow (by Daniel Kahneman)

Websites:


data, coding

What does a data scientist look like? My journey to becoming a data scientist and a mother

data, coding

By Karin Sasaki, Senior Consultant in Data Science, Ekimetrics

Career paths are rarely straightforward, are they?

When I was studying, I wanted to become an applied academic, using data to solve problems. I completed my PhD in Biomathematics, which led to my first job: working for the European Molecular Biology Laboratory. I was helping to create mathematical models of biological systems.

Six years later, and I’ve pivoted a little from the original life plan. I’m now working as a data scientist for Ekimetrics, where we use data to help businesses better understand consumers, their marketing, or to improve their products and services.

It was a leap of faith to leave academia, but I had found a true passion in data; I love the different ways it can be used. And so, maybe unsurprisingly, I landed on data science as a career path. And so, the research began, and I read deeply and widely, looking for areas that interested me and ways to get a foothold in the industry.

Read. That’s one key piece of advice I’d give to anyone worried about switching careers or trying something new. Really throw yourself into the literature around a subject and spend free time learning more about it. Most things aren’t as terrifying or as difficult to understand as you might expect!

For me, I found a really great data science community online – but I’d say the same is true across a great many industries. People write really helpful how-to articles, and they’ll offer to help you find the answers when you need them.

Connecting with people and networking is another great way to find out whether you’re comfortable in a certain field.

The journey rather than the industry

I’ve always been ‘subject agnostic’ and more interested in the process of finding an answer, rather than a specific sector or industry.

So, to me, a company like Ekimetrics with many different clients and types of businesses is fascinating. I love being able to use my background to take data sets and translate them into something businesses can understand and feel comfortable with.

Marketing effectiveness is particularly interesting to me because there are so many different data types and customer interests to analyse. At the moment, I’m involved in a couple of projects that are helping companies understand customers and their behavior better. In turn, this is helping us outline the most valuable areas of each business, so we can see how it can better serve its customers.

It’s rewarding to see our work have a direct and measurable impact on the success of a business. And it’s brilliant to feel we’re helping in a tangible way.

One Tech World Virtual Conference 2022

01 APRIL 2022

Book your place now to what is becoming the largest virtual conference for women in technology in 2022

FIND OUT MORE

Choosing to become a mother

Of course, when you love your career, motherhood isn’t always a straightforward choice to navigate. The fear of losing your career can be a daunting. I am sure countless women have struggled with this and I know it still influences women’s choices around the world.

I’m very grateful to my family, friends, and work colleagues, both male and female, who have supported me. I’m now blessed to have a son.

When I was pregnant – and then again when raising a small child – I noticed the support of those around me the most. At work, my colleagues saw my output wasn’t diminished, I was just working in different patterns.

Regardless of how I got the work done, they still trusted that projects would be finished on time and to a high standard. Having this trust motivated me to get the job done well. My husband has been wonderful and taken on more around the house whenever I needed to interview for roles or work particularly hard. And having my extended family to help gives me more time to develop my career.

Ultimately, while it should always be a woman’s choice to become a mother, it isn’t always possible to do this alone. Parenting itself is often a full-time job and so support is vital to continue to thrive and reach your potential in the workplace.

Diversity within Science, Technology, Engineering and Maths (STEM) careers

Obviously, my thoughts on juggling children and work are probably less relevant to much younger girls and women who are thinking about a STEM career!

In data science, like any industry, there is no one way in. There are many different routes and I am proud to demonstrate that!

To anyone who is, and is feeling unsure, I studied maths because I was interested in it, but I have seen that if your interests change you can change your career plans too. You’re never pigeonholed into something for life. If you want to make a change, go for it!

I did a lot of studying online and there is more support than ever, for example, via e-courses. Plus, being a data scientist doesn’t mean you’ve had to have a particular background in maths. Actually, in line with removing bias, it is good to have a wide number of backgrounds in any particular team. Different viewpoints are welcome.

Having a mix of people in terms of gender, academic and cultural backgrounds, changes the dynamics of a workplace for the better. It means being able to bring your full self to work and not be afraid of expressing yourself. In a workplace like that people feel freer to be and express themselves and that positivity permeates into the work and collaboration.

It also creates better business outcomes. According to the McKinsey study “Why Diversity Matters,” companies in the top quartile for gender-diverse executive suites were 15% more likely to generate above-average profitability compared to the bottom quartile of companies whose executive teams were predominantly white and male.

I hope that women such as myself can continue to break the bias around certain careers and encourage diversity. In doing so, I truly believe we will support much more prosperous societies and do better business.

Karin SasakiAbout the author

Karin is a mathematician with a Ph.D. and five years’ experience in modelling and data analysis in various industry and academic settings. She has worked with a variety of data that has come from molecular biology systems, as well as from operational research, and now marketing. Her specific modelling and analytical skills include low dimensional topology, topological data analysis and machine learning.