fbpx

Learn about the role of data science in fraud detection at Sendwave

4.+2022+CP+Blog+thumbnail+resized+(14)

ARTICLE SUMMARY

In the first of our series showcasing the exceptional female talent of Sendwave, we hear from Kerri Rapes - Director of Data. She gives us invaluable insight into Sendwave’s mission and values as well as guidance for transitioning to a support role, developing your career and seeking out relevant resources to keep learning.

What is your background?

In the past, I worked as a Field Data Specialist at Schlumberger in the deserts of Oklahoma. Seven days a week I lived on oil drilling locations collecting, verifying, and interpreting large amounts of raw data from the borehole. Being the only “technology savvy” person for miles I became the computer specialist, the directional drilling co-pilot, the communications expert, and the cell phone repairman.

Eventually, I decided to leave the company and pursue Machine Learning, a topic which I had then been independently studying for over a year and a half. I needed more challenges in my life, and I wanted to be able to exercise my creative side in impactful (and admittedly more ‘earth-friendly’) ways. I moved to South America and began leading a team of 8 individuals focused on proposing, implementing, and maintaining Machine Learning solutions. Eventually, our work at Conversica led to two patents and the successful sale of the startup.

Afterwards, I was part of the founding team at Odd Industries; I collaborated with a team of 9 engineers to build deep learning and autoML models, create platform architectures that support automatic retraining, and assess data quality with automatic detection of drift and poor segment performance.

And what do you do at Sendwave now?

When the pandemic rocked the world, the construction focused startup (Odd Industries) started to lose steam and I took an opportunity to lead the anti-fraud program at Sendwave. The experience was thrilling and as the needs of the company changed, I slowly moved into my current position as owner of the Data job family and director of both the anti-fraud and data engineering teams.

I’m currently working to develop our data program. I spend a lot of time questioning what parts about our system cause it to be slow or difficult to make data driven decisions and then working through roadmap options that would better position us. Currently I’m heavily involved in:

Risk

Our Anti-fraud team. This team leverages Data Science to identify and stop bad actors from committing fraud on our platform.

Data Foundation

Our Data Engineering team. This team is working hard to bring different (previously siloed) data sources together and build the capabilities like data quality checks, data reconciliation, data modelling, BI support, ML modelling, and more to enable other teams to be more data driven.

What about the Sendwave mission resonates with you the most?

I’m an immigrant myself. I moved from the United States to Chile 6 years ago and understand the pain of needing to frequently move money from one country to another. I would consider myself someone who is “financially aware” and hate that I have to pay an “Immigrant Tax” just to be able to use my own money. It makes me feel good inside to know that we are forcing into question these practices that large institutions have had in place for so long.

How do Sendwave’s business values shape how you deliver your data science and fraud detection work?

We see a lot of ugly things on the anti-fraud team. It’s easy to slip into a mentality that everyone is guilty until proven innocent. Sendwave’s business values are obviously the exact opposite of that. There’re a lot of cross-team conversations that happen within engineering that while being sympathetic to the woes of fraud, we’re insistent on the innocent until proven guilty mindset. This ultimately manifests itself in the form of monitoring edge cases and not imposing authoritarian rule onto the product.

You’ve recently become Director of Data. What advice would you give to others transitioning into a leadership role? 

Uff … ok, [Small beads of sweat start to form on her forehead, haha!] I think “leadership role” is kind of a loaded phrase. Being a “leader” is disconnected from being placed in a given role. A leader is someone who is respected and fills a mentor-like space. It emerges slowly over time through consistent upholding of values like dedication, reflection, patience and comradery.

What I have recently transitioned between are roles. Moving from Data Science Manager to Director of Data really just means I moved from supporting one single team to several teams. I understand my contribution to be a facilitator for the “builders” actually solving the problems and making this whole thing work. In short, move blockers and provide resources to allow the developers to create the best solution possible.

My advice for anyone transitioning into a supporting role is to set up a regular habit of self-reflection. You are not better or above those that directly report to you. You are a teammate filling a need in the organization. Doing honest check-ins will help to ensure that you’re collaborating, not controlling and that you are providing direction via trust and not obligation.

What tools, resources, or community organisations have you found useful when developing your career?

I paid for University with student loans and graduated absolutely overwhelmed with debt. That experience shaped my whole approach to education going forward. I love online learning and Massive Open Online Courses. Some are better than others of course and I find those that charge a modest tuition are generally more beneficial that the completely free tier. Some of my favourite resources are: Udacity, AWS certification courses, Google Cloud Platform certification courses, and most things produced by Andrew Ng.

What are your top remote working tips?

Make time for the chit-chat. Working remote it’s easy to be very strict about calendars and meeting agendas but if all anyone ever talks about is work, a lot of the sense of community one feels when working in a team is lost. I believe it’s hugely beneficial to allow a few minutes to talk about life in each meeting or schedule regular coffee meetups where the goal is to talk about anything EXCEPT work.

And a little bit about you….

How do you spend your time when you’re not at Sendwave?

Hiking, skiing, sea kayaking, white-water kayaking, surfing, diving. I took a three-month road trip through Patagonia a few months ago and before that I spent 6 months road tripping from Tennessee to California and back.

What was your first job? 

I worked as a local Carny. I grew up in a small town and the parents of my sister’s friend owned a small set of traveling carnival rides. When I was 16, I spent the summer working in all the local fairs for minimum wage. I wouldn’t recommend it; it was very hot with a lot of sun exposure and angry drunk parents. However, it was an experience! I feel like I’ve tried all the fair food that exists (like fried Kool-Aid for example) and enjoyed dancing in the evening with strangers 😀

How do you define career success?

I love what I do and hope to continue to evolve with this budding industry in the future. I define success in the work environment as being placed in a position where I’m surrounded by people that want to fuse science, technology, and intuition, with a little faith, hope, and dedication.


Learn more

RELATED ARTICLES

Explore the journey of Mitra Goswami, Senior Director of Data Science & Machine Learning at PagerDuty, in building inclusive AI products. With a background in...
Darya Petrashka, data scientist, share her tips on data science training for women in tech.
Darya Petrashka, Data Scientist, shares her top certification picks for data scientists.
Darya Petrashka, Data Scientist, shares how to find a data science mentor and how to build upon the relationship.

This website stores cookies on your computer. These cookies are used to improve your website and provide more personalized services to you, both on this website and through other media. To find out more about the cookies we use, see our Privacy Policy.