From cutting-edge machine learning to the complex realities of fraud prevention and experimentation at scale, this is a behind-the-scenes look at how data science is solving today’s toughest challenges – and shaping tomorrow’s opportunities.
Hosted by the brilliant team at Deliveroo, you’ll hear short, fast-paced talks, network with fellow data practitioners, and leave energised with fresh ideas and inspiration.
The Line-up
Ella Johnson-Watts – Staff Data Scientist, Deliveroo
Session: High-quality experimentation at Deliveroo: The vision, implementation and reflection
Over the last decade, Deliveroo has built a culture of running high-quality experiments. This talk reflects on Deliveroo’s experimentation journey, from the initial vision to what we have achieved so far with some key lessons for low-cost democratised experimentation at-scale.
–
Dr Dimitra (Mimie) Liotsiou – Senior Research Data Scientist, dunnhumby
Session: Causal inference at dunnhumby: how we are applying causal methods in retail data science, key tips and learnings.
It is well known in data science that ‘correlation is not causation’. However, standard data science and machine learning methods are about correlation, not causation. Therefore, to answer questions about cause and effect (e.g. about measuring impacts, uplift, or about why a KPI moved), which are central to many data science projects across sectors, a new, causal, approach is needed, of which the standard data science methods are just one part. In this talk, I will show how we are using the new science of graphical causal inference at dunnhumby to answer cause and effect questions in retail, and I will share key tips and learnings.
–
Lydia Monnington – Data Science Lead, Google Deepmind
Session: Advanced SQL – top tips for dealing with real world data
Level up your SQL knowledge to deliver results in real world situations.
I’ll cover:
- Window functions, what makes them great and an example of how to use them to create sessions.
- Arrays, why data engineers use them and how to access data stored in them easily.
–
Susan Blaszczak – Staff Data Scientist, Deliveroo
Session: Fraud Prevention at Deliveroo – the role of Science and key challenges.
Fraud prevention is a significant challenge for a platform like Deliveroo and this talk is all about how data science can help. I’ll talk about some of the key fraudulent behaviours we try to tackle and dive into the tricky parts – balancing stringent fraud prevention measures with a seamless user experience and the inherent difficulty in establishing ground truth for fraudulent behaviours.
Please note that the recording for this session is not available.
Supported by
