12-05, 16:30–17:00 (UTC), Data/ Data Science Track
Taking any project from zero to production is challenging. And Data Science has a particularly high failure rate, with a lot of ideas not getting beyond the prototype stage.
But there are real reasons for this: there is intrinsic and unknown complexity in data, and there are often big challenges knowing if we have actually solved the problem -- the answer is so rarely "yes" or "no".
In this talk I'll cover some key learnings from a decade working on DS problems at early- and later-stage startups, building products to improve product market fit.
Taking any project from zero to production is challenging. And Data Science has a particularly high failure rate, with a lot of ideas not getting beyond the prototype stage.
But there are real reasons for this: there is intrinsic and unknown complexity in data, and there are often big challenges knowing if we have actually solved the problem -- the answer is so rarely "yes" or "no".
In this talk I'll cover some key learnings from a decade working on DS problems at early- and later-stage startups, building products to improve product market fit.
I'll cover insights from combining Product Management and Data Science, the benefits of being part of customer discovery processes as data scientists, and how to identify and reduce the biggest uncertainties--whether they're in our models or in the real world.
I'll also discuss some of the traps I've fallen into over the years, how (hopefully!) to avoid them, and finally some patterns that have let me learn and iterate quickly towards solving real company and customer needs.
No previous knowledge expected
I’m a product-focused data scientist who’s been part of the startup scene in London for over a decade.
I’ve been responsible for shaping products for the UK and US markets from an early stage, building models and infrastructure in the finance space, and authoring fairness policies to ensure models are both compliant and ethical.
I try to spend as much time as I can on understanding the problem. It’s usually a lot messier and more complex than I realise, but it’s key to building solutions that have a real impact!