Scotiabank Global Banking and Markets
Platform Product Lead, DevOps, Cloud and Machine Learning · 2015 to 2016
Situation
Global Banking and Markets wanted to ship software faster and put machine learning to work. No Canadian bank had a production application on public cloud, and the bank's own infrastructure was provisioned and changed by hand.
What I did
Designed the bank's first fully automated infrastructure platform, secured the funding, hired and trained the DevOps team, and migrated GBM's technology teams onto it. Secured the security and compliance approvals for fully automated production deployments to the data centres, AWS and Azure. Ran the platform as a product, with the development teams as its customers. Consolidated the division's four global Kondor+ trading systems into one. Shipped the first production application on public cloud at the bank, a machine learning risk application, and the deep learning credit card collections tool with DeepLearni.ng, covered by the Wall Street Journal.
Result
Infrastructure support effort down 90%. Deployment frequency up 640%, to 4,550 changes a week. When AWS opened its Canadian region, Scotiabank was running in it within 90 minutes, which AWS announced itself. From 2016 the approach went bank-wide: see Scotiabank enterprise agile and DevOps adoption.
What it took
Getting risk, security and audit to agree the controls once, in the pipeline, so no team had to negotiate them again. Teams adopted the platform because it made their week shorter, and the numbers followed.