Development Deployment Delivery of Data Products
Brilliant ideas alone have never made any firm successful. It is the spot-on implementation of these ideas that matters. In data science, ideas manifest themselves into data insights; and implementation turns these into Data Products.
YQ
Machine learning offers a fantastically powerful toolkit for building useful complex prediction systems quickly ... it is dangerous to think of these quick wins as coming for free. Using the software engineering framework of technical debt, we find it is common to incur massive ongoing maintenance costs in real-world ML systems...
D. Sculey et al. (Hidden Technical Debt in Machine Learning Systems ) Google
Machine learning offers a fantastically powerful toolkit for building useful complex prediction systems quickly ... it is dangerous to think of these quick wins as coming for free. Using the software engineering framework of technical debt, we find it is common to incur massive ongoing maintenance costs in real-world ML systems...
D. Sculey et al. (Hidden Technical Debt in Machine Learning Systems ) Google
Implementation does not, however, come out of the box. Achieving the security, operational and reliability standards of data products pioneered by the DevOps world, yet internalizing the fit-for-purpose codebase of data scientists and ML engineers, is not a simple quest. There are a number of all-mighty black-box data science platforms promising seamless integration of your data insights into day-to-day production cycle if only you committed to their rules of the game. YQ is none of these…
YQ is a solution built on strong foundations of production-grade, open-source technologies that allow your data scientists to choose their favourite tool without worrying about the subsequent integration hell. We are completely open on which open-source technologies we use under the hood.
Kubernetes-native
Cloud-agnostic
Focused on financial
industry
It goes without saying, we do not charge for packaging an open-source tool stack. You only pay for the value added coming from our service – reliability, security, best practices and day-to-day operational excellence.
“The obvious conclusion: If you’re interested in enterprise IT infrastructure, Kubernetes should be your technology of choice...”
Jason Bloomberg, SiliconANGLE
The obvious conclusion: If you’re interested in enterprise IT infrastructure, Kubernetes should be your technology of choice...
Jason Bloomberg, SiliconANGLE
Development
Do you appreciate your data scientists?
Give them a tool they actually like to work with. YQ does NOT force them to use proprietary platforms with a steep learning curve and expensive trainings; nor libraries that are outdated before the platform update is ready to be rolled-out.
YQ integrates the open-source data-science champions…
Turning ML scripts into production-grade apps, with all the system, library and data dependencies is time consuming. Organizing these into robust workflows is complex; and maintaining these can become daunting.
Using cutting-edge technologies, YQ helps you to deploy your ML models before you lose your head and they lose their edge.
Containers
Continuous
integration
Version control
Deployment
Delivery
In batch or real-time, large or small data, complex or simple workflows, manually triggered or automatically scheduled, YQ makes no excuses when it comes to reliable delivery process.
Even more, YQ silently collects delivery-oriented metadata from that are made readily available in the monitoring dashboards.
Batch / real-time
Monitoring
dashboards
Orchestration