Software Coffeehouse

by foxteck

A blog of Coffee-driven Software Engineering.

Rapid experimentation with Large-Language Models

A fast feedback loop for improving LLM accuracy on Data Extraction. Evaluation results for GPT-4o, Claude Sonnet and o1-preview.

A post about re-architecting queueing tasks with serverless for improved processing speed, error handling, observability and traffic spikes.

A post about queueing theory, Nudge scheduling and Simulation.