Improving the auditability of AI interactions is crucial for maintaining security and control. A recent update focuses on preventing the exposure of raw code to AI models, enhancing data security, and providing better insights into flagged code changes.

The Challenge of Raw Diffs

Previously, raw git diffs were sent to AI models for analysis. This approach, while providing detailed context,

This post discusses a recent update to our post generation process, shifting from hardcoded sections to a more adaptable, example-driven approach.

The Old Approach

Previously, our post generation system rigidly enforced a predefined structure: Introduction, Development, Examples, and Conclusion. This inflexible format ignored the nuances and varied structures of different example posts,

Introduction

This post addresses an issue where a billing system, specifically when using a service like Cashier, fails to correctly identify active subscriptions beyond the default type. We'll explore the problem and provide a solution to ensure all subscription types are properly recognized.

The Problem: Default Subscription Type Bias

In many billing systems, especially those utilizing