Webinar Wrapup: How Self-Isolation Affects Data Science Work

Unique Challenges of a Remote Data Science Project Team

It’s difficult to argue that data scientists should be working under the same roof when a highly contagious pathogen is continuing to spread, especially with today’s remote work capabilities. But that doesn’t mean it’s easy or that data scientists should expect anything resembling “business as usual.” Working from home presents significant hurdles, such as staying connected with colleagues and navigating distractions at home.

The top challenges for data scientists are communication and collaboration, according to Van Loon. This includes the rapid deployment of new software and means of working together. “This seamless collaboration of being in the same room is gone now,” Van Loon said. “Now it has become a real obstacle. For data scientists, this situation is only intensified.”

Even though it wasn’t unusual for data scientists to work remotely before the pandemic, they always had a more concrete infrastructure to fall back on. Now, in-person collaborations are no longer possible in most cases. It’s much more difficult now to replicate experiments, recreate test environments, and perform other processes typically done in one location. 

Other challenges stem from the difficulty of shifting to an at-home work situation, such as the need for a dedicated workspace, managing distractions at home (especially if you have children), and staying motivated. 

“This is being recognized by many authorities as really a major barrier for data scientists,” Van Loon said. “Teams working from home might be in different time zones, they might have different schedules, and they might be dealing with their partners or their kids.” 

Of course, data scientists also need operational reliability, security, access to data, and the ability to use the most effective tools of their trade. When each team member is using a different network (even if they’re ultimately logging onto the corporate VPN), reliability, speed, and security can vary. Also, synching work through email or Google Drive, for instance, adds another step and can slow things down, Van Loon explained.

How Isolation Impacts the Roles of Team Members 

“Many data scientists are feeling isolated, they’re feeling lonely, disengaged, sometimes unmotivated, or unable to switch off from their work,” Van Loon said, adding that barriers to collaboration can cause a great deal of stress. The key to a successful transition to the current, remote work landscape is proper planning without engaging in micromanaging, he explained.

This is accomplished by setting broad team goals and clearly stating expectations—trusting that your team will get the job done—while refraining from taking a one-size-fits-all approach. You want to help keep everyone on the same page, but it’s important to realize that different team members will have specific needs and schedules, depending on their particular work-from-home situation. 

Van Loon suggested the following solutions to current challenges pertaining to time, team management, and productivity:

• Agree on a dedicated “turn off” time in order to help workers draw a line between work and personal life, essentially preventing burn-out
• Set concrete priorities with measurable benchmarks 
• Create a check-in schedule for each workday
• Consider scheduling team-building events via online video 
• Encourage the use of online video to conduct brainstorming sessions
• Share actual work progress, not just code, with team members
• Use a version control system
• Understand that everyone has a different work environment, while encouraging team members to establish a structure for communication and deliverables
• Focus on maintaining the team’s core culture
• Make sure data access and sharing is democratized and consider the use of digital sandboxes

Best Communication Tools for Maintaining Team Interactions

None of the goals and suggestions mentioned above are achievable without the right tools when face-to-face interactions are all but impossible. Van Loon prefaced his suggestions with the reminder that everyone has their own preferences, needs, and security requirements. What’s most important, he said, is to “structure your work, try to face each other, and have a good process to communicate, to plan.” 

Van Loon’s top suggestions for tools to communicate and streamline workflows include the following:

Zoom

Zoom is far from the only video conferencing platform available, but its advantages include to easily screen-share, record sessions, use the whiteboard function for brainstorming sessions, and integrate with other popular tools (including Slack).  

Comet

This online marketplace for freelance engineering and data professionals allows organizations to upload their assignments and receive qualified candidates within 48 hours. 

Slack

This popular collaboration tool allows for real-time messaging, archiving, granular search functions, and transfers of relatively large digital files, while allowing a wide range of integration with other collaboration, communication, and monitoring tools.

Domino

This enterprise-grade data science management platform is optimized for AI modeling and integrates with a host of existing tools used for tracking, reproducing, and comparing experiments.  

Trello

Trello is a collaboration tool that integrates with many other popular applications, similar to Slack, that helps you organize your projects into “boards” and gives you a complete view of a project’s progress, who’s working on what, and so on.

Staying Productive and Maintaining Growth While You’re Isolated

When your team is together, working under the same roof, synergy comes easily. But when everyone is stuck inside their home and dealing with different obstacles, demands, and routines, productivity can take a hit. Maintaining professional—and personal—growth also can be difficult, especially when you’re not out there networking with other data science professionals.      

“Humans like routines. But I think there are things people can do to alleviate their pressures and turn their remote work experience into a more positive and possibly more predictive one then perhaps what you had at your office,” Van Loon explained.

Keeping the same routines you had while working in the office is not reasonable or even possible in most cases, but Van Loon suggests certain measures that can help keep your team productive without burning out or hindering personal growth: 

• Maintain a structured schedule
• Keep a regular sleep schedule, work during set hours, and set aside time for socializing and non-work activities
• Take up a hobby, take time to cook a nice meal, or make sure you engage in activities that make you happy
• Set achievable goals, allowing for “small victories”
• Pace yourself

The overriding theme is balance and routine, which will help you stay focused when it’s time to work without feeling overwhelmed or stressed. To this point, Van Loon suggested that team check-ins shouldn’t only be about work or the project at hand, but also socialization and meaningful connections that otherwise would be reinforced around the office water cooler or during lunch breaks. 

“Humans are social creatures,” he said. “Just socializing will go a long way toward motivating yourself and your colleagues.” 

Looking Ahead: Data Science Work After COVID-19 

While acknowledging what’s lost when team members can’t work directly together, Van Loon points out some of the positives of the current work-from-home situation, which include ditching the commute and spending more time with family members. He said he expects the future of data science work (as well as other fields) will be a hybrid of pre-COVID-19 work processes and elements of remote work, something “more optimal, more efficient, then what we’ve seen in the past.” 

During the global pandemic, he explained, a period of digital transformation that normally would take a few years is taking place in a few months for many organizations. It’s challenging and potentially stressful, but will generate a lot of useful insight about how to move forward once we get back to what our “new normal” will be. 

As we emerge from the pandemic and its workplace disruptions, Van Loon expects to see the following impacts on data science work in particular:

• Organizations will make it a priority to develop data science best practices, as well as remote work best practices in general
• Companies will learn how to be more flexible, adaptable, and accessible during future disruptions and to account for more remote work
• A greater focus on clear metrics for success
• Greater adoptions of streaming analytics, cloud infrastructure, and advanced natural language processing (NLP) tools  

View Simplilearn’s webinar with Van Loon in its entirety, titled “How Self-Isolation Affects Data Science Work.” 

Stay Positive, and Productive, in Our Data Science Career

The exciting field of data science was already in a state of massive growth prior to the pandemic, but the disruptions that have occurred as a result have only bolstered its importance across most industries. Data scientists are in demand; are you prepared for the future? Get ahead from the comfort of your own home and seize the future today, with Simplilearn’s award-winning online bootcamps. We have several Data Science courses in collaboration with IBM.