← 🤖 Retrieval & AgentsNEXT IN RETRIEVAL & AGENTSContext Engineering for Agents→
Core
Agent Design Patterns: ReAct, Plan-and-Execute, Reflection
These are the named control-flow architectures for LLM agents: ReAct interleaves reasoning and actions in a tight loop, plan-and-execute decomposes the task up front and then runs the steps, and reflection adds a self-critique pass that revises output. Each trades latency, token cost, and robustness differently. Applied AI interviews probe this to see whether you pick a pattern from task structure rather than defaulting to one loop for everything.
a free account unlocks the core curriculum tier · no card
RELATED CONCEPTS
PRACTICE THIS IN REAL QUESTIONS
RAG & Agent System DesignWhat is the Plan-and-Execute agent pattern, and how does it compare to ReAct?→LLM & GenAI FundamentalsWhat is ReAct (Reasoning + Acting) prompting, and why does interleaving thought and action help?→RAG & Agent System DesignWhat are Tree-of-Thoughts and LATS, and when is search-based planning worth the cost?→System Design for AI in ProductionDesign a deep research agent that answers complex questions by searching and synthesizing many sources.→LLM & GenAI FundamentalsWhat is in-context learning, and why does chain-of-thought prompting improve results?→LLM & GenAI FundamentalsExplain self-consistency and tree-of-thought prompting. When are they worth the extra cost?→
COMPANIES THAT ASSUME THIS