From the course: Data Science Team Lifecycle Management

How to onboard new employees

From the course: Data Science Team Lifecycle Management

Start my 1-month free trial Buy for my team

How to onboard new employees

“

- Onboarding new employees may seem straightforward, but few people have a repeatable scalable system to do it efficiently. Let's outline the major components of a good onboarding framework. A sound onboarding system is comprised of three key areas. The business overview, the technical overview, and the people overview. Taken together, any new hire will have an understanding of their place in the organization, the relationship with other teams and departments, and how what they work on contributes to the overall growth of the business. Let's start with the business overview. A business overview has a few essential things to understand. The first is a high level understanding of what the company does and how it makes money. If you're working in a non-profit organization, this would be the mission and how that mission is funded. Once you outline how the business works overall, the next drill down level is the division or team level. Every company is different, so focus on each division or team's contribution to the overall company. For example, a data scientist working for an online real estate website might work on home valuations. In contrast, another data scientist in a different part of the company may work on home recommendations, so on and so forth, but they all work towards a common goal. Next, the technical overview. This is the most significant portion of the onboarding process and requires the most time. Here you'll want to cover at least these six topics. One, data systems and architecture. Two, machine learning models currently in production. Three, code repositories and CI/CD pipelines. Four, product documentation and knowledge repositories. Five, the machine learning production life cycle, including how data flows into and out of the organization. And six, ticketing and prioritization systems. Finally, the people. There are many ways to approach this, including traditional org charts, which are helpful, but org charts don't always tell the whole story. Personally, I think it's good to highlight any interpersonal dynamics that new hires should be aware of. I don't mean gossip, I mean objective observations on how well teams function. That kind of candor will help to build trust more quickly. There are three core dimensions of personnel every new hire should understand. I think of these as center, up, and out. Center, their role in hierarchy relative to others on the immediate team. Out is their immediate peer group relative to the other groups, for example, the relationship between marketing and sales. And up, who are the people above them? What do they care about? And how and when to engage them? Lastly, and don't forget this, if you're a data science manager or have direct reports, your responsibility is to help new hires navigate the organization successfully. And so there you are. Now you know a basic, repeatable onboarding framework, and you can use that for any new hire, regardless of hierarchy or role.

Contents