Niki Agrawal’s passion for inspiring creativity with the scientific method led her to her career as a Data Science Consultant with Pandata.
Read on to learn more about her journey and how it’s shaped her career in data science.
I was a kid who loved to learn.
From dance class to science labs, I was interested in a wide range of subjects, and enjoyed employing both analytical and artistic thinking to solve problems.
I grew up in a very creative household, so I had always felt comfortable expressing my love for the arts—but I was also very intrigued by the clarity and structure of math and science.
My search for a field of study that would allow me to combine my strengths ended at Case Western. Their biomedical engineering program offered the framework of a scientific education while also strongly encouraging interdisciplinary collaboration and creative problem solving.
I later attended New York City’s Data Science Academy, which solidified the machine learning, computing, and analytics skills I first acquired while working in research.
After graduation, I spent the first few years of my career in healthcare technology. It was (and still is) an exciting space, and I really enjoyed building technological tools to help patients manage their health.
I’m very passionate about translatable tools and wanted to apply my data science expertise to a range of problems.
Pandata’s clients represent several industries, including healthcare, finance, and technology, and the consulting team celebrates very diverse backgrounds. I enjoy this variety, so data science consulting seemed like the perfect next step for my career.
Data storytelling is a key part of my work. It takes creativity to put yourself into the shoes of the client or end user and understand which insights will be the most relevant. At the end of the day, my main concern is: How can I use data to tell a story that is compelling to my client and creates true business value?
My overarching goal is to work with end users to recognize and solve problems.
Each aspect of that process ultimately comes down to communication. My success hinges upon how clearly I can pin down different applications of the same data set and share those use cases with a client. Data visualization, for example, is a key area in which I get to present insights creatively.
The most difficult part of my job is working through “bad” data to produce a functional output. There’s a phrase in data science—“garbage in, garbage out”—that stresses the importance of teaching machine learning algorithms with clean, unbiased data. Balancing all of the moving parts of this process can be tricky, especially as you’re working to simultaneously manage client expectations.
I often see the fear that AI will replace human jobs in all sorts of industries, and it’s critical to dispel that notion. In fact, it couldn’t be further from the truth!
Holding ourselves accountable to creating explainable AI models helps a lot, too. Our clients appreciate knowing exactly what’s going on at any given time in the design and development phase.
AI refers to a broad set of technologies, including machine learning and natural language processing, enabling computers to learn tasks that typically require human intelligence.
New accessibility models are on the rise with tools and platforms that promote low- to no-code solutions. If you can understand the math, you can apply data science without getting caught up in the logistics of coding.
Opportunities to implement AI are vast and growing. However, the usefulness of these insights is often limited by access to “good” data.
Business leaders should prioritize having a strong data strategy in their company, so that new AI solutions can build upon trusted data that is relevant to the business problem at hand.
Data science is constantly evolving. It helps to connect with data scientists outside of your company to stay engaged and up-to-date on applications in other industries.
If you can make data science a personal and professional interest, you’re a lot more likely to stay inspired.