Workshop: Prompt Engineering with Watsonx
summary
Hands-on walkthrough of Watsonx Prompt Lab that shows live edits beating guesswork: swap models (flan_ul2, Llama 2, StarCoder), raise max tokens to avoid cutoffs, and add a simple one-shot prompting example to turn repetition into a useful numbered list. Demonstrates decoding modes (greedy vs sampling), temperature and random seed for reproducibility, repetition penalty and stop sequences, and mentions a free trial plus ten lab exercises—small models (20B) behave very differently from a 70B Llama 2.
AI-written summary · about
chapters (22)
- 0:00Workshop introduction
- 1:05Lab purpose and structure
- 3:18Accessing Watsonx and Prompt Lab
- 5:03Structured vs free-form views
- 6:28Free-form prompt interface
- 7:11First prompt attempt and output
- 8:56Choosing the right model
- 11:43Tokens and model parameters
- 13:12Adjusting Max tokens
- 15:06Improving prompt structure
- 17:14One-shot prompting with examples
- 21:16Refining output with parameters
- 22:20Decoding methods impact
- 23:25Repetition penalty and stopping
- 25:10Temperature and creativity
- 27:16Random seed for reproducibility
- 28:41Moving to Lab 2 exercises
- 29:23Exercise: creative sentences
- 31:25Model guardrails and examples
- 33:16Model size effects
- 34:22Advanced prompt structuring
- 36:05Wrap-up and next steps
related talks
59mTutorial: How to Win Friends & Influence LLMs (with Prompt Engineering)
26mHow To Win Friends & Influence LLMs (with Prompt Engineering)
01h 35mBuilding with Anthropic Claude: Prompt Workshop
01h 39mMastering LLMs: From Prompt Engineering to Agentic AI
13mPrompt Engineering Simplified
39mOptimizing a Prompt for Production
26mMastering Large Language Models: Prompt Engineering for Precision & Performance
30mIs Prompt Engineering Dead? How Auto-Optimization is Changing the Game