Data Science roles have unrealistic expectations from blog posts and other job descriptions. If you are going to copy one I hope you pick this one!
You are probably not an expert in Data Science. That’s okay, the fact that you are reading this puts you above other companies in the search for a Data Scientist.
**What is a Data Scientist?**Data Science is an ocean as deep as it is wide. A Data Scientist is somebody who can:
- Communicate their problems and decisions.Program.Visualize and investigate data.Shape a dataset and manage storage and transformation.Make meaningful predictions using a dataset.
That list is in my priority order. Machine learning is the least important aspect of Data Science in industry, though it’s still fundamental.The above is a big ask. The title Data Scientist currently represents the jobs :
- Project ManagerSoftware EngineerData AnalystData EngineerMachine Learning Engineer
Common MistakesHere I will go over some of the common red flags that show me you don’t know why you are hiring a Data Scientist.
PhD onlyThe industry is currently divided between Data Science as a product, and as research. A few rare companies join these teams.A team focused on research and creating new novel solutions to problems has the need for PhD holders.According to Indeed, 20% of Data Scientists have a PhD, of those a minority have a PhD relevant to the field. It is not the PhD that makes these data scientists useful but their scientific approach. By requiring a PhD you are saying no to 80% of the Data Scientists that currently have experience. Scientific reasoning has a place in Data Science. I would choose industry experience over a PhD every time.
Focus on Machine LearningMachine Learning is foundational to Data Science. Without good communication Machine Learning never delivers value. Without the ability to deploy a model to production you are creating work for a DevOps team.Unless Data Science is the core product of your company you are going to find value applying third party products. Google, Amazon, and Facebook all offer tooling that requires skill to install. There is no need to reinvent the wheel. Machine Learning ability is useful for answering questions about a model’s validity.
Missing Software EngineeringSome companies will not allow Data Science to be part of the tech arm of their business. Data Science is a technology role. Good Software Engineering practices will make your Data Scientist’s products a pleasure to integrate.
Low SalaryIn Data Science both the company and candidate have high expectations. Machine Learning is a popular dream industry. We aren’t hiring Machine Learning specialists, we are hiring Data Scientists.As described above, a Data Scientist is somebody who is doing many jobs. My graduate Data Science salary was £25,000, the same pay any graduate software engineer got. If you want somebody as multi-faceted as a Data Scientist you need to put your money where your mouth is. Even more so if you are expecting PhD holders.For your first hire, you need a top Data Scientist to build up the business function. For this type of person, you are looking at around £60,000 in Manchester. This Data Scientist will have the role of Business Transformation added to the pile. Their job is to change perspectives on how the business can use data.What I often recommend is to hire either one senior or two juniors that fill in the gaps in each other’s skillset.
#data-science #hiring #data #recruitment #towards-data-science #data analysis
