This site uses cookies. By continuing to browse, you agree to our use of cookies as outlined in our Privacy and Cookie Policy.
One of Europe’s leading media companies had a strategic ambition to personalise its content across a variety of platforms and channels, and required an MLOps platform to enable it to maximise ROI from its data science projects. Its Content Discovery team worked closely with Datatonic to fast-track its MLOps maturity to be able to deliver customer value faster and with flexibility, while embedding its reliability engineering principles throughout. This new MLOps capability allows it to offer increased levels of personalisation to its customers, frees up Data Scientists’ time, and promotes a culture of experimentation for impactful Machine Learning use cases.
Built an MLOps capability to deploy and monitor Machine Learning models in production with reduced risk, less effort and faster time to production.
Deployed 2x Machine Learning models for automated tagging and ranking of content, to increase the personalisation of individual customer experience across a range of OTT services.
Improved the productionisation, optimisation, and automation of Machine Learning models with MLOps best practices using GCP and open source tooling (such as Kubeflow Pipelines).
Personalisation in any industry is about delivering the right message at the right time to the right consumer. In the case of a large company, this requires analysing massive amounts of data across many sources and connecting them correctly to the individual user.
To maximise the value of their data, our client needed to empower teams to build, own and operate ML capabilities in production, and address the manual processes involved in deploying them.
Automation of Machine Learning processes is key for enabling models to keep up with the fast-changing, new data from customers. MLOps ensures that ML models built by the team: (1) do not deteriorate, (2) are easier to maintain, and (3) can improve and adjust on their own.
“From a business standpoint, being able to get models into production faster and with less risk increases the engagement and the revenue that you can drive from those models.”
To help our client achieve personalisation at scale, the Datatonic team built an MLOps platform with best practices and tools to take its ML models into production with the goal of facilitating greater personalisation.
During the engagement, Datatonic’s MLOps experts worked closely with the Content Personalisation team to:
This work enabled our client to more easily develop, deploy, and productionise their Machine Learning models, and the increased automation throughout the process allows Data Scientists to focus on more complex work.
If there’s data, we can optimise, innovate with and automate it to drive business value. Find out how we’ve supported the biggest brands across industries.
Know exactly where and how to start your AI journey with Datatonic’s
three-week AI Innovation Jumpstart *.
* Duration dependent on data complexity and use case chosen for POC model
With your own data sets, convince your business of the value of migrating your data warehouse, data lake and/or streaming platform to the cloud in four weeks.
With your own data, see how Looker can modernise your BI needs
with Datatonic’s two-week Showcase.