Data science engineer: A day in the life


A common narrative for a day in the life of a data scientist is that we build the next cutting-edge machine learning (ML) model, present it at conferences, and enjoy the applause. However, this is far from the daily reality for most data scientists.

In reality, data science is much more pragmatic: instead of building new models from scratch, we refine known models from known libraries and select predictors based on domain knowledge. We can quickly create added value from this.

To debunk the myths, here’s my career as a principal data science engineer at revenue intelligence firm Xactly:

My daily chores

The daily tasks of a data scientist often depend on the size of the team. In larger teams, the roles tend to be more specialized. There will be multiple people working on a specific phase of the data science lifecycle. In a smaller team like mine, you have to wear multiple hats and understand all the phases.

[ Also read Data scientist: A day in the life. ]

My day starts by asking my teammates about the different models they are working on or the problems they are solving. Sometimes all is well, and we celebrate that. But usually there are numerous challenges to overcome.

After connecting with my team, I work on my models and projects, which tend to be more difficult and complex. I try to dedicate the majority of my time to solving challenges ranging from improving model performance to solving deployment issues.

Also Read :  Phillies' Kyle Schwarber breaks Statcast with mammoth NLCS Game 1 home run off Yu Darvish

Show the value of data

Between communication, data engineering, meaningful results reports and more, data scientists have many goals. At Xactly, my daily goal is to demonstrate the value of our data to the rest of the organization and our customers.

Strategy and evangelization are a big priority. It’s important to illustrate how data science is useful in other departments such as engineering, marketing, customer experience, and sales. In a day, this can get messy and force us to delve into the details of data creation. We hope to use this to create new predictors that could be incorporated into our models.

My team focuses on solving various technical issues across the organization on a daily basis. Over time, daily work helps to achieve greater goals. I see it as solving one or two sub-problems a day, which over time feeds into the solution of a larger problem that serves a larger purpose.

As we complete projects, we build on that success by developing new models and gaining new insights. For example, a recently deployed model achieved nearly 100 percent sales forecast accuracy. Now we’re building the same ability into our other models.

How I accept hybrid work

Xactly seems to encourage a hybrid working model. So when I have days focused on collaboration and communication, I find it more effective to work face-to-face with my team and stakeholders.

Also Read :  The Art and Science of Fitness | The role of exercise in treating depression

[ Related read Remote work: How to balance flexibility and productivity ]

On immersive tech days when I don’t need a distraction, I typically work remotely. Remote work is more productive to dive deep into data and solve complex problems. Math, analysis, and software engineering are core components of my work-from-home approach, and the fewer distractions the better.

Data science is constantly evolving

For data scientists, the challenge is not necessarily building models or writing code; The key question is what models to build and what code to write.

A lot has changed since I got into data science in the 2010s. Most importantly, expectations are higher – and we have automation in place to meet those expectations.

In the past, the great expectation of data scientists was to create a model. That alone was often enough to keep stakeholders happy, let alone explainability. There was no need to answer questions to justify the predictions, because the idea of ​​building an ML model that spits out predictions was downright futuristic and promising in itself.

Over time, however, this raised questions about the accuracy of the predictions. This led to an influx of explainability frameworks that show the impact of model predictors. This is where evangelistic responsibility begins to come into play.

Also Read :  DJI Osmo Mobile 6 Pushes Smartphone Photography Further

Eventually, increased automation and community support came along, paving the way for data science to reach new heights. But that leads to another misconception, which is that at some point there will be no need for data scientists because automation will replace us.

I do not agree with that.

For data scientists, the challenge is not necessarily building models or writing code; The key question is what models to build and what code to write. The challenge is to find the best model that delivers the greatest business value. And to find that out, you need business knowledge that AI doesn’t have.

Data science is forecast to grow faster than almost any other field by 2029. The data science profession is not going away anytime soon. So if you are considering a career in this industry, I would suggest blocking the misconceptions surrounding it. My advice is to find a project idea that is of great interest to you and build it from scratch. Combine this hands-on experience with some education via online courses and books, and you’re on your way to a successful data science career.

[What is a day in the life like in your role? If you’d like to participate in this series, reach out here.]



Source link