Transforming the Retail Experience with Cloud Foundry and Machine Learning - Michael Goddard
summary
A microservices demo running on Cloud Foundry on GCP links a Python Twitter adapter through Spring Cloud Dataflow and Spring Boot processors to Google Cloud Vision, Natural Language and TensorFlow-based models to detect positive sentiment and identify dresses, match against a crawled catalog of ~4,000 images, and generate a GPS-linked promo code for 30% off in-store; outbound offers appear within ~10–15 seconds in the demo. Code is on GitHub and the pipeline is stateless and autoscaling.
AI-written summary · about
chapters (9)
- 0:00Introduction and setup
- 1:27Project background and retail challenge
- 4:57Demo: Personalized retail offer
- 9:34Cloud Foundry and GCP architecture
- 13:13Twitter data processing pipeline
- 16:26Machine learning APIs in action
- 21:51Image matching with TensorFlow
- 25:37Open source and future steps
- 26:20Q&A and closing thoughts
related talks
28mPivotal Cloud Foundry, Google Machine Learning, and Spring
22mOperationalising Data Science on Cloud Foundry
26mWriting Cloud Native Applications to Execute on Cloud Foundry
01h 06mOrchestrating Data Microservices with Spring Cloud Data Flow - Mark Pollack
36mCatch Me If You Can – Leveraging Cloud Foundry, SDN, BOSH and Predictive Analytics
49mDDD Strategic Design with Spring Boot
27mBuild Intelligent Business Applications with Machine Learning and Cloud Foundry
04mDistributed Cloud for retail and manufacturing