PinnedIsaac Tonyloi·Sep 14, 2024End-to-End Machine Learning on Google Cloud with Dataflow and BigQuery Using Open DataIntroduction
PinnedIsaac Tonyloi·Sep 1, 2024Understanding Cost Functions and Their Role in Measuring the Accuracy of Machine Learning ModelsIn the world of machine learning, the accuracy of a model is important. But how do we quantify how well a model performs? This is where the…
PinnedInThe StartupbyIsaac Tonyloi·Jan 31, 2021Supervised and Unsupervised LearningWhat is machine learning?
Isaac Tonyloi·Oct 29One warehouse, six audiences- Branch-level analytics for a Kenyan bankThe problem with one set of dashboards
Isaac Tonyloi·Oct 26Building a credit risk report for a Kenyan bankI built the report a credit risk team puts together every month, for a made-up mid-sized lender I called Mwangaza Bank.
Isaac Tonyloi·Oct 25Why Contoso stopped growingContoso is an electronics retailer with 74 stores across eight countries and a busy online shop. Its revenue nearly quadrupled between 2017…
Isaac Tonyloi·Oct 15Britain after coalOn 30 September 2024, Ratcliffe-on-Soar power station in Nottinghamshire shut down for good. It was the last coal plant in Britain, the…
Isaac Tonyloi·Sep 23Engineering for Regional Erasure: What AWS me-south-1 Teaches Engineering LeadersOn September 15, 2026, AWS told customers that data held only in its Bahrain region (me-south-1), or only in one UAE availability zone…
Isaac Tonyloi·Sep 20Prompt injection defenses, actually red-teamed and scored, not just describedAsk almost any team building an LLM agent how they handle prompt injection, and you’ll get an architecture diagram, input sanitization…
Isaac Tonyloi·Sep 18Enforcing data contracts in CIA producer team renames a column, or quietly changes it from optional to required, and nothing breaks on their end. Their own pipeline runs…