I've been playing with various ML training tools and different monitoring and operations tools. I've been unsure if it's one size fits all (e.g. langsmith or mlflow) or whether certain tools are more proportionate for the need. I haven't gone into licencing costs but for each ML Type, I have put together a list of… Continue reading MLOps Toolsets for Different ML Types
Setting up MLOps for repeatable pipelines when using scikit-learn Not every AI problem requires a large language model. In many enterprise environments, the most valuable systems are still well engineered, explainable, repeatable and operationally governed. This is where classical machine learning pipelines still provide benefit and Scikit-learn remains one of the strongest foundations for these… Continue reading MLOps for Scikit-learn
Federating Data Between Snowflake and Databricks with DuckDB and Apache Iceberg If you're running both Snowflake and Databricks — and most enterprises I work with are — you've probably hit the federation problem. Data lives in both platforms, analysts need to query across them, and the obvious solutions (ETL everything into one place, or pay… Continue reading Ducks on Icebergs
The rule of law is a fundamental principle from the Mesopotanian Code Ur-Nammu, through Magna Carta to International Criminal Court's decisiion to ditch Microsoft Office for European open source alternatives. Data sovereignty requires certainty that services will never be terminated or at the mastery of a governmental body. For this reason I find it useful… Continue reading EU Sovereign Cloud List
Faster Insight, Better Reuse, and Scalable Data Foundations Healthcare organisations face growing demand for better use of data: improving operational performance, supporting population health management, enabling AI, and accelerating research. Yet many still rely on fragmented pipelines, duplicated transformations, and slow bespoke data requests. At the same time, the economics of technology have changed. Modern… Continue reading Explaining Medallion Data Architectures in Healthcare
Medical data ownership needs to pass from the trust to the patient to enable ML diagnostics and better research collaboration. The best way of achieving this without being compelled by the hyper-scalers is to move to a persistent container based architecture.
As a Google Cloud Platform certified architect I really should blog some more about my actual usage of GCP. One of my favourite tools is Dataproc as it provides a managed Spark & Hadoop environment and enables a lambda architecture suitable for complex network event processing and function remediation. A mobile radio network is a… Continue reading Google Cloud managed Spark & Hadoop for Mobile Network Performance
Implementing a successful 5G deployment means moving to a micro-service architecture. To support the major technical use case of 5G it is important to decouple the micro-service architecture from the data fabric. This article describes some of the key 5G use cases requiring a data fabric.
Monitoring and observability across hybrid clouds is complex and requires avoid stove pipes by working in a multi-cluster environment.
This is a fantastic list of Cloud Native networking technologies. Because it is curated it is updated frequently. It is making my life much easier. https://jimmysong.io/awesome-cloud-native/ And the tutorial list for old guys is especially useful