A local open source proof of concept that puts a semantic layer between a language model and an Iceberg lakehouse, using Strands, WrenAI and Trino.

Dynamic DES v0.8.1 adds data lake integration, so one SimPy codebase writes batch Parquet data for ML training and streams live Kafka events.

An Apache Flink and Kotlin streaming architecture where online machine learning detects concept drift from machinery wear, controlled by a shadow mode router.

Apache Flink, Kafka and Valkey turn a contextual bandit recommender into an event driven service with stateful training and low latency inference.

A Python prototype of a contextual multi-armed bandit recommender that simulates user behaviour and validates the algorithm against cold-start users.