best practices

Use LLMs to get insights from disorderly data

This document publishes hard-earned best practices on how to use large language models (LLMs) to extract reliable insights even when you have poor metadata

Slow is smooth, and smooth is fast - Part 1

Part 1 of how a software delivery team used a data-driven approach to remove friction and deliver faster. It lays the backdrop and expands on the friction in the system

Slow is smooth, and smooth is fast - part 2

Part 2 of how a software delivery team used a data-driven approach to remove friction and deliver faster. It details on the technical, organizational and process improvements made to resolve the issues