Tuesday night, Miner & Kasch partnered with the Colorado Data Science Meetup to host a panel discussion on women in data science. Three of our top data scientists, Julia Maddalena, Karen Farbman, and Jordan Hagan came together — in a night of candid conversations — to discuss the challenges that women in tech and data science face.
The discussion explored trends in tech and a range of day-to-day issues people encounter in their data science careers.
A prominent theme of the night was imposter syndrome — the tendency to doubt your skills or accomplishments, and to fear that you’re always on the verge of being exposed as someone who doesn’t really know what they’re doing. In The Imposter Phenomenon in High Achieving Women: Dynamics and Therapeutic Intervention, Pauline Rose Clance and Suzanne Imes note that imposter syndrome “appears to be particularly prevalent and intense among a select sample of high achieving women.”
High achieving women in data science are no exception. One female audience member described how despite earning a PhD in Aerospace Engineering, she did not feel smart enough for a job in aerospace or data science. But there are ways women can help each other to overcome the effects of the imposter syndrome. As our panelist Karen Farbman noted, meetups like this one are a great opportunity for community members to lift each other up. Insecurities can arise constantly at one’s job or in everyday life, and being in a room with successful, like-minded women can help others voice their concerns and deal with them effectively through a supportive environment.
Another topic of debate explored the day-to-day challenges of data scientists, including how to work with management on data science projects and how to convince business people what you’re doing is important.
Julia Maddalena explained the fallacy of building the perfect model. She shared her experience of putting too much emphasis on developing the flawless model. Instead, she said, a strong focus on the business problem should be a data scientist’s first priority. Jordan Hagan noted “It is often beneficial to get a proof of concept end-to-end system up and running and then go back to refine your model.” Building a minimum viable product exposes a lot of roadblocks quickly, so starting those conversations early not only ensures a smoother build process, but also keeps managers up-to-date on your progress.
In retrospect, the night was full of meaningful and open conversations. An engaged audience picked our brains about what a day in the life is like … best practices for real world applications … and getting your career off the ground. Anything was fair game! And with this attitude, we opened the eyes of many attendees to foster a world of data science that is an inclusive and diverse environment ready to build the next great data products.
