Work
Selected work
Real problems. The thinking behind them. The decisions that shaped the architecture.
Each write-up covers one engagement: the problem, the decisions, and the architecture that followed.
Case studies
SLM or LLM? The More Interesting Question Came First.
Instead of immediately comparing models, I stepped back and looked at what the platform actually needed the AI to do.
Read the case studyBefore Building the Factory of the Future, Find the Data Problem
When a manufacturing organization begins an Industry 4.0 transformation, the obvious questions are usually about machines, IoT, AI, dashboards and automation.
Read the case studyEngineering ANT
ANT is a reusable platform for building governed AI workers: agents that reason, connect to enterprise systems, and execute business workflows.
Read the case study
Next
Have a decision forming?
If a technology or AI commitment is taking shape and the direction still needs pressure-testing, start with a conversation.