Interview Questions

Rishabh P. - "Step 1 — Clarifying questions (always start here) Before jumping to solutions, ask these three things: What type of product is it? Open-ended chatbot vs. task-oriented assistant? An open chatbot has the lowest risk appetite because the user can input anything. Assuming: it's a shopping assistant (like Rufus on Amazon). Is it stateful or stateless? A stateful chatbot stores memory across sessions, which makes PII handling critical. Assuming: Stateful What kind of safety are we"See full answer

Shalin G. - "Tell me about the most recent product you launched that you are very proud of?"See full answer

S R. - "For RAG systems, You need to evaluate the retrieval and generation. Typically, you have golden truth question and answers as evaluation data set. For retrieval, check if the retrieved contexts are relevant for the question For generation evaluation, You check the semantic similarity between the golden truth and RAG generated answer. Apart from this, you can evaluate the output using frameworks like RAGAS where the answers generated are evaluated based on completeness, faithfulness, to"See full answer

Rohith K. - "Stakeholder management is kinda a critical aspect of product management. Because you would be the glue between the leadership team, design, GTM and engineering team. My simple approach to handle different stakeholders is - Deeply grounding myself into the product vision and business value that we are envisioning to create. Any decision or initiative needs to directly or indirectly benefit that path Open communication - any new feature, solution or product decision taken needs to be communi"See full answer

Tracy M. - "Clarify LLM purpose; research, extraction, reasoning, planning, assessing? Assuming document extraction skill only, large db, multiple doc types, unstructured data; assuming Haiku-2.5 (less complex tasks) and LLM parsing tool exists pre-extraction call. Assuming golden dataset for ref is accessible, experiments and observability set in place. Likewise prompts. Eval set:document extraction coverage: % of accurately extracted (doc type, structured outputs) document field/value/format coverage"See full answer