Back to AI Agents / RAG
About this interview
A technical interview on Agentic Design Patterns, pitched at the medium level. A voice AI interviewer leads the conversation, adapts its questions to your answers, keeps you on topic, and afterward gives you honest, specific feedback on where you were strong and where to improve. Expect roughly 30 minutes.
What you'll be assessed on
Distinguish the three memory scopes — working (scratch-pad), short-term (session), long-term (persistent) — and their appropriate storage backends
Explain how episodic and persona memory enable personalization across sessions
Describe practical implementations such as in-context history, vector stores, and knowledge graphs for memory retrieval
Identify when memory grows stale or bloated and strategies for pruning or summarizing it
Topics covered
Memory fundamentalsWorking memoryMemory typesLong-term memoryMemory backendsSession and working memoryPersona memoryEpisodic memorySelf-improvement through memoryEntity memoryLong-term memory via toolsIn-context vs external memoryMem0 memory pipelineKnowledge agent pattern
A few sample questions
Just examples to set expectations - the real interview has many more and adapts to your responses.
“In everyday terms, what does it mean for an AI agent to be stateless, and what problems does that create for users?
“What is the knowledge-agent pattern for self-improvement? Describe the role of the side agent that observes the main conversation.
“How do you handle privacy and data governance for long-term agent memory? Consider a user who wants their history deleted — what does that require technically?