Context Engineering & Prompting with Sari Itani скачать в хорошем качестве

Context Engineering & Prompting with Sari Itani

In this session from our Pre-Winter Sprint Bootcamp (January 24), Sari Itani explains the difference between prompt engineering and context engineering, and why writing better prompts alone is not enough to get reliable results from large language models (LLMs). This talk breaks down how prompts work, why vague prompts lead to weak outputs, and how techniques like few-shot prompting and chain-of-thought help guide model reasoning. Sari then introduces context engineering as a way to build intelligent systems around LLMs by controlling what information the model sees and when it sees it. The session also covers key ideas behind agentic AI systems, retrieval-augmented generation (RAG), memory, tools, and context safety, showing how modern AI systems go beyond simple prompting to become structured decision systems. Topics covered in this session: • What prompt engineering is and why it matters • Few-shot and chain-of-thought prompting • Why prompt engineering has limits • What context engineering means • How agents, retrieval, memory, and tools work together • Context windows and cost vs performance • Safety, bias, and data leakage risks in LLM systems This talk is useful for: AI engineers, prompt engineers, machine learning students, developers working with LLMs, and anyone interested in building reliable AI systems. Main takeaway: You don’t just write better prompts, you design the context the model sees. 📍 Recorded during the Pre-Winter Sprint Bootcamp – January 24 🎥 Watch and share with anyone learning about prompt engineering, context engineering, or LLM systems. Slides available at : https://www.canva.com/design/DAG_UIpe... #PromptEngineering #ContextEngineering #LLM #ArtificialIntelligence #GenerativeAI #RAG #AIEngineering #MachineLearning #AgenticAI #Bootcamp