# @amazonwebservices on Instagram

- **Type:** Video
- **Original URL:** https://www.instagram.com/p/DNwRLM90oWq
- **Gondola URL:** https://gondola.cc/posts/33747328-amazonwebservices-instagram
- **Thumbnail:** https://img.gondola.cc/tr:w-,h-,fo-auto/postThumbnails/052108b068.jpg
- **Posted:** 2025-08-25T12:00:05.000+00:00
- **Account Owner:** AWS (@amazonwebservices) — https://gondola.cc/amazonwebservices

## Caption

⭐️ Watch as darkosubotica shares 3 tips for working with LLMs while enjoying a Balkan breakfast 🥐☕️

1️⃣ Prompt Engineering: Craft your prompts carefully, be specific, give clear instructions. It's an art - get creative with some examples or try chain-of-thought prompting 🧠
2️⃣ Output Parsing & Validation: LLMs may not always return what you expect so have robust error handling. Check out this Python code example for guidance (link in bio 🔗) 
3️⃣ Model Selection & Fine-tuning: Choose the right model for your use case, don't overshoot. Fine-tune for accurate, formatted responses tailored to your needs 

Implement these best practices using #AmazonBedrock 🚀

## Stats

- **Views:** 6,038
- **Likes:** 244
- **Shares:** 0
- **Comments:** 12

## Tags

amazonbedrock

---
Copyright (c) Gondola