DevOps at ING Analytics

on YouTube ↗43m

" summary

ING moves its analytics from batch to continuous delivery using an engineering culture, standardized pipelines (Jenkins, GitLab) and chapters/tribes to ship a real-time decision engine built on Apache Flink and Kafka with ML models deployed via PMML. The pipeline detects event patterns (CEP), scores models for relevance, and pushes personalized propositions into the banking app; it runs in production, stops a fraud attempt and drives show/accept/success funnels. The claim: DevOps plus streaming reduces time to value and increases operational reliability.

AI-written summary · about

" chapters (7)

1. 0:00The importance of data
2. 3:38ING's cultural journey
3. 7:48DevOps and engineering culture
4. 12:29Continuous analytics model
5. 16:54Case study: decision engine
6. 20:34Technical architecture
7. 26:57Q&A and concluding remarks

" related talks

• 46mFast Data at INGBas Geerdink, Martijn Visser · Flink Forward Berlin 2017
• 27mData Analytics Democratization: How ING Data Analytics Platform Bootstraps New...Krzysztof Adamski · Open Source Strategy Forum London 2021
• 44mA Tale of Two Clouds: Analytics Platforms at INGJohn Müller · GOTO Amsterdam 2023
• 38mTaking away customer friction through streaming analyticsFerd Scheepers · Flink Forward Berlin 2017
• 50mData Science in the EnterpriseBas Geerdink, Natalino Busa · J-Fall 2015
• 51mReal Time Investment Alerts using Apache Kafka & Spring Kafka at ING BankMarcos Maia, Tim van Baarsen · Spring I/O 2019
• 26mApplying DevOps practices in Data and ML EngineeringAntoni Ivanov · Conf42 Kube Native 2023
• 51mFrom DevOps for data to DataOpsVincent Goris, Niels Nagle · SQLBits 2024