Context engineering is the new prompt engineering.
And it’s becoming the most critical AI skill.
Together with @MiqJ (Product Lead at @OpenAI) we created a comprehensive guide.
Key insights: 🧵👇 https://t.co/SkvC37K1Ns
1. What Is Context Engineering
It is the art and science of building systems that fill LLM context window to improve their performance.
Unlike prompt engineering, context engineering is a broader term with many activities that happen also before the prompt is even created. https://t.co/fmoswANtsM
2. Types of Context
There are 6 types of context:
- Instructions
- Examples
- Knowledge
- Memory
- Tool results
- Tools https://t.co/djLM7lBDe5
I prepared two examples you can analyze (GitHub):
- PM Agent: https://t.co/gDI1gRtioz
- Lovable Bug Fixing Agent: https://t.co/aiJGvArvYg
Before we continue, I recommend his AI Product Management Certification. It’s a 6-week cohort taught by Miqdad Jaffer.
The next session starts on Sep 15, 2025. A special discount for our community: https://t.co/uDiI5HZsSr https://t.co/Uk2HGZAiKN
3. What Is RAG and Where It Fits In
RAG is often labeled a context engineering technique. But RAG is a three-step pipeline:
- Information Retrieval: Pulling data from external sources (e.g. vector DBs, APIs)
- Context Assembly: Structuring and filtering the retrieved data into a prompt
- Generation: Using an LLM (or agent) to generate the output
Context engineering doesn't include generation.
4. Information Retrieval Techniques
You can see a free RAG simulator: https://t.co/jv4UBasLUz
5. Context Assembly Techniques
Aim to provide minimal, relevant, and well-structured information.
Good context engineering requires a mix of retrieval and assembly techniques. https://t.co/uH366FgR6h
@n8n_io example: https://t.co/VxRDsedEU0
Find this helpful?
See our full post: https://t.co/zYpLkBwd0g
Enjoy this?
- Follow me @PawelHuryn for more
- Share this thread with others
I appreciate it!
https://t.co/FSds57ePX7
