Everyone is an expert and each one is allowed to do their own part separately but when we come together the magic happens. The team lead, along with Math and Stats person, the Data Visualisation, Storytelling team and domain experts all work in unison. This according to Ashwini Mathur, Head of Data Science at the Novartis AI Innovation Hub in Dublin is essential to delivering great data science.

This was the rhythm of guest contributor Ashwini guest talk on last weeks #AIFightsBack webinar series. He gave insight into how to become a Data Science company instead of a company with Data Scientists.

The art of data science

It's clear that at Novartis this is driven from the top down, with the CEO Vas Narasimhan big on how they are becoming a medicines and data science company. This according to Ashwini, is where it becomes much more about the company culture and the thinking around this than just the discipline of good data science. This is the art of using data science throughout the whole organisation.

Video

Slides

How to Become a Data Science Company instead of a company with Data Scientists - tales from Novartis from Ruth Kearney

Talking the talk

Everyone in the company has to understand data science talk, from senior managers to sales teams to lab researchers. It’s much more about culture and thinking than about hiring and upskilling individuals. Novartis have an “unbossed” culture driving “curiosity” and “inspiration”. Ashwini outlined the four major drivers in the company to achieve this;

1. Open environment
2. Learning culture
3. Innovation mindset
4. Support for risk taking.

The last one will see the launch of the new AI Innovation Hub in Dublin this Autumn. It’s a strategic partnership with Microsoft where they will collaborate on developing drugs with AI. The Novartis campus in Basel, Switzerland and the Global Service Centre (GSC) in Dublin will work with Microsoft’s Research lab in Cambridge. Two companies who are masters at what they do join forces to solve healthcare problems have the capability to disrupt healthcare. A partnership that requires both to be driven by data science.

This journey to become a data science company requires a strong leadership, an open culture, an appreciation of data science talk throughout the company and innovate partnerships to accelerate transformation.

This all helps manage scientific humility and makes sceptics of all of us, which according to Ashwini makes the world a better place to live in.

#AIFightsBack AI

Matt Macdonald-Wallace (Mockingbird Consulting) and Dr. Iain Keaney (Skellig.ai) look at the use of IoT and privacy respecting data science to help businesses operate in the post-COVID19-lockdown world.