Scaling a Machine Learning Social Feed with Feature Pipelines

apply(conf) - Apr '21 - 10 minutes

In 2018 we launched an experiment to add machine learning to the ranking algorithms on the social feed of the Cookpad application. The results of this experiment were plausible for our users, however the architecture we built for this experiment did not allow us to scale beyond a limited number of users. Therefore, in our next iteration, we focused on redesigning the architecture to scale to our global user base keeping in mind all the learnings from our first experiment.

In this talk we will discuss why a feature store is essential for serving machine learning at scale. We will describe the feature store solution we have built, its architecture and the pipelines populating the feature store. Finally, we will discuss the optimisations made to our feature store in order to serve data for online inference in our production environment.

Ettie Eyre

Platform Engineering Lead


Cookpad is the world’s largest recipe sharing platform, with over 100 million users using the platform each month in over 75 countries.

Ettie joined Cookpad in 2018, and built a Machine Learning Infrastructure Team to empower research engineers to productionise machine learning capabilities. She moved to a new role as Platform Engineering Lead in 2021.

Prior to joining Cookpad she spent 10 years building machine learning systems across a range of startups after completing a PhD at Bristol University in AI .

Ettie co-organises Bristol Machine Learning.

Nadine Sarraf

Machine Learning Engineer


Nadine joined Cookpad MENA (Beirut,Lebanon) as a Machine learning researcher in 2016, a subsidiary of the Japanese largest recipe sharing service, where she learned and applied AI to real-world problems.

Nadine was a City AI ambassador in Beirut, Lebanon in 2018; a global applied artificial intelligence community where she facilitated knowledge exchange by mapping the local AI ecosystems, conducting talks about applying artificial intelligence and sharing her personal experience to enable more people to learn and collaborate in the field of applied AI.

She relocated to Cookpad’s headquarter in Bristol in 2019 where she continues integrating AI into Cookpad’s platform.