Hardcoding LLM prompts is fine until it isn't. Here's what we built instead.

I had a bug last month that took most of a Saturday to find. A support bot we shipped started...

martedì 30 giugno 2026 New tab

635 words~3 min read

I had a bug last month that took most of a Saturday to find. A support bot we shipped started promising refund timelines that didn't match policy. Customer complaints, frantic Slack messages, the usual.

The prompt had changed three weeks earlier. Nobody could remember why. Git blame pointed to a one-line edit inside a 200-line SYSTEM_PROMPT constant. No PR description, no diff worth reading.

That's when I knew I'd been writing prompts wrong for the last two years.

PromptOT - Prompt Management Platform

Compose prompts from typed blocks, version safely, and deliver to your apps via API. The prompt management platform built for AI engineering teams.

Hardcoding LLM prompts is fine until it isn't. Here's what we built instead.

Hardcoding LLM prompts is fine until it isn't. Here's what we built instead.

Related reading

Hardcoded System Prompts: An Anti-Pattern in Production

Prompts as Code: How to Version, Test, and Ship the Prompt Layer in 2026

Your LLM Prompts Are Running Ungoverned in Production. Here's the Architecture…

How I A/B test LLM prompts without fooling myself

Prompt versioning: how I learned the hard way

How a model upgrade silently broke our extraction prompt (and how we caught it)

Related reading

Hardcoded System Prompts: An Anti-Pattern in Production

Prompts as Code: How to Version, Test, and Ship the Prompt Layer in 2026

Your LLM Prompts Are Running Ungoverned in Production. Here's the Architecture…

How I A/B test LLM prompts without fooling myself

Prompt versioning: how I learned the hard way

How a model upgrade silently broke our extraction prompt (and how we caught it)