Keeping up With Data — Week 9 Reading List
5 minutes for 5 hours’ worth of reading
Who would have guessed that one day I’ll be attending a Machine Learning conference in Prague from Zurich? Well, it happened last weekend. Obviously, it’s a different experience, no small talks and networking during coffee breaks and no beers in the evening but I was really impressed by the event. I was able to spend the weekend with the family, watch a couple of talks live and few others from recordings. As with WFH and the office post-pandemic ‘dilemma’, the future of conferences is likely to be a hybrid between physical and virtual. And, I actually think it’s probably a good shift.
But for now, let’s hit it the reading road:
- The 3 Missing Roles that every Data Science Team needs to Hire: “Data science is a team sport.” And while FAANG interviews are focusing on statistics, coding, machine learning and who-knows-what, successful teams need a bit more than a group of algorithm experts with amazing coding skills. The roles needed are — (1) data science translator to ensure the right business problems are being tackled and they are correctly understood; (2) behavioural psychologist to capture the ‘people element’ of the data-powered solutions; and (3) data storytellers to infuse life into data insights and make the solutions compelling. Many data projects are failing not because the algorithms are wrong but because they don’t solve the right problems. (Ganes Kesari @ Towards Data Science)
- The best HR & People Analytics articles of February 2021: I don’t even know how I came across this newsletter but after quickly scrolling through I actually ended up clicking on most of the links. It is amazing to see how data and people are interconnected topics. Data professionals have realised that whatever they do will, in one way or another, by consumed by people — and HR is using data to make better decisions and automate operations. I’ve even noticed that data scientists are moving to HR and vice-a-versa. Let’s hope that data will help people be both productive and happy at the work. (LinkedIn)
- Writing at Work — The Why, What, How Framework: Some data professionals consider writing anything but code a waste of time: “Just tell me what you want and let me work.” And then, to the surprise of all involved, we often realise we misunderstand one another. In software engineering they separate the roles of people writing the specs and user stories and those coding the solutions. But I’d argue that in data science, due mainly to the non-linearity and R&D nature of the process, it is important that everyone is able to clearly communicate the understanding and proposed solutions. Writing is therefore an important skill for data professionals. One-pagers for project prioritisation, scoping documents and after-action reviews are a fantastic way to add structure to the otherwise complex process of data science projects. (Eugene Yan)
I took part in another virtual event last week where experts from IBM’s elite data science team spoke about building and scaling AI with trust and transparency. I really enjoyed the talk and a follow-up discussion about bias (and fairness), drift and explainability. But one thing that really caught my attention was how many people in the US can be uniquely identified just by a birthdate, ZIP code and gender. Keep this in mind when you’d like to suggest that anonymising data by removing name, emails and likes is enough.