Sumanas Sarma

Machine Yearning: A Detailed Study of Taking ML to Production

Artificial Intelligence (AI) and Machine Learning (ML) are all the rage right now. In this session, we’ll be looking at how (at Sainsbury’s Argos) we took an end-to-end case from R&D to a production microservices environment. We will cover a) a specific business case, and how we built it, b) some engineering best practices that can be applied to ML, c) how ML research can be integrated with an agile development cycle, and d) how open ended research can be managed within business strategy. Additionally, we will address some of the typical problems we encountered and overcame for example: a) confidence in what only few people understand b) ways of working between Data science and IT c) ways of working between business, data science and IT and d) the actual development process. At Sainsbury’s Argos we are trying to go from a simple idea (and sometimes a business situation) to a production ML system. Along the way we had to integrate open ended academic research tasks with our existing agile development process and project planning, as well as working out how to deliver the ML system to a production setting in a repeatable, robust way, with all the considerations expected from a normal software project.

Sumanas works on taking deep learning models from an exploratory R&D stage to production. He has a total of nine years experience with a Masters in Software engineering. Apart from working on JVM-based languages (Groovy, Java, Scala) he has experience with Haskell, Python and Javascript. Using CNNs (and more recently GCNs) on NLP challenges is his current focus.

Buttontwitter Buttonlinkedin
This website uses cookies to ensure you get the best experience. Learn more