M R (@mrober_) profile picture
Joined March 2025
Photo by M R (@mrober_)
Always implement fallback parsing for AI model responses! We just fixed a critical bug where markdown code fences (json) broke our income predictions. - Clean markdown first - Add regex fallback - Never assume perfect JSON Here is a prompt to implemment this:
2
32
... 1. Strips markdown code fences first 2. Tries normal JSON parsing 3. Falls back to regex extraction if JSON fails 4. Handles edge cases gracefully Show me both the cleaning logic and regex fallback patterns."
16
"I have an AI model that returns JSON responses, but sometimes they're wrapped in markdown code fences like json or javascript. This breaks my JSON.parse() calls. Can you help me create a robust parsing function that: ...
22
Anthropic released their complete prompt engineering guide. 99% won't read it. I spent 3 days testing every single technique. While you're saying "please" and "thank you" to AI, smart operators are using advanced techniques to 10x their output. Here's how you can do the same:
60
337
15
3,139
678,137
Claude Code Pro Tip: "Ultrathink with deep analysis and comprehensive planning: 1️⃣ First explore the codebase and gather full context 2️⃣ Use sub-agents strategically where they add value 3️⃣ Plan thoroughly before implementing Then explore, plan, and implement the following:"
1
143
This is an example that I used to ask to refactor a classifier script.
33