Imaginary Test Data. Real Token Bill.
Testing AI applications with invented traffic looks fine until real users arrive. Then come the retries, the fallback models, and the token bill.
Posts by Alan Mon
Alan Mon is a contributor to the Speedscale blog. • 9 posts published
The AI Code Explosion: Why Your Mocking Strategy is Breaking Down
AI-generated code breaks traditional mocking. Here's how BYOC capture and replay with proxymock keeps verification grounded in real production traffic.
The DIY Traffic Replay Trap: Why AI Scripts Fail at Scale
AI-generated scripts seem like a quick win, but DIY traffic replay tools become a maintenance nightmare at scale. Here's why Speedscale beats the prompt.
When Your Observability Literally Stops Traffic
Observability tells you what failed—but not how to recreate it. Why reproducibility is the missing fourth pillar, and what that means for incident response.
Prompt, Deploy, Pray Is Dead: Validating AI Code with Proxymock
AI-generated code is moving fast—but without behavioral validation, you're gambling with production stability. See how Proxymock changes the equation.
The Ultimate Stress Test: Kim Jong Un, a Parachute, and a Multi-Billion Dollar Prop Bet
From memes to market meltdowns: Explore how a multi-billion dollar Super Bowl prop bet involving Kim Jong Un pushed prediction markets and SRE teams to.
The CES Hangover: 3 Expensive Hardware Fails That Were Actually Software Problems
Stop letting third-party API spikes crash your app. Learn why mocking latency for OpenAI, Anthropic.
Digital Twins Gone Wild: My Unexpected AI Doppelgänger
Tired of Staging environments that don't match Prod? Learn how Speedscale's Digital Twin, built from real production traffic, solves dependency.
Shopify Outage 2025: Rise of the Commerce Kaiju
Shopify's infrastructure crumbled in 2025 like a city under a 'Commerce Kaiju' attack.
Try Speedscale free
Choose the desktop proxymock or the hosted cloud trial to get started.