My Local LLM Scored 6/6. It Was Wrong Every Time.
Six months of trying to make a 1.2B model useful, and the measurement mistakes I made along the way.
Read articleLocal models. Real tools. Honest measurements.
Building AI systems that earn their confidence.
I’m Mark Hall, an AI engineer at GoEngineer. I write about local language models, agentic systems, evaluation, and the engineering between a convincing demo and dependable software.
Latest writing
Notes from the workbench
Open source project
Sir Thaddeus
A local-first AI assistant for Windows with permissioned tools, durable memory, visible evidence, and a hard stop button.