Questions 1–200 of 274
How would you design and implement a production RAG pipeline?
Asked in 19 of 942 interviews · last asked Oct 20 · Design · Mid-level
Show answer
RAG retrieves relevant documents and gives them to a language model to help answer a question. I would parse documents, split them into chunks, and store embeddings with source details and access rules. For each question, I would combine keyword and vector search, rerank results, and generate an answer with citations. If evidence is missing, say so.
- Use RAG for changing or private facts; fine-tuning mainly changes model behavior, not access to current facts.
- Enforce document permissions before sending context to the model; treat retrieved instructions as untrusted.
- Track document versions, refresh changed content, and remove deleted content from indexes.
- Test retrieval quality and answer accuracy with known questions; monitor latency, cost, and failures.
What the interviewer checks: The interviewer checks whether you can build a secure, reliable retrieval system and measure answer quality, not just connect a model to a vector database.
Other ways it was asked
- “What is retrieval-augmented generation (RAG), and when would you use it?”
- “What is retrieval-augmented generation (RAG)?”
- “How can you optimize a retrieval-augmented generation (RAG) system, and why might you choose RAG over another approach?”
How would you design and implement a generative AI application end to end?
Asked in 4 of 942 interviews · last asked Sep 9 · Design · Mid-level
Show answer
I would define the user task, success measures, and privacy needs first. Then I would choose a model and build the flow: authenticate requests, fetch needed context, call the model, validate outputs, and return results. I would test quality and safety before release. In production, I would monitor quality, latency, failures, and cost.
- Use retrieval for private or changing facts; enforce access rules before sending context to the model.
- Compare models and prompts on realistic test cases, including wrong answers and unsafe requests.
- Keep secrets out of prompts; treat retrieved text as untrusted and check permissions before tool actions.
- Set timeouts, retry limits, rate limits, and fallback behavior; release gradually and support rollback.
What the interviewer checks: The interviewer checks whether you can turn a model call into a secure, reliable application with measurable quality.
Other ways it was asked
- “How would you explain the flow of an end-to-end generative AI project?”
- “How would you explain a generative AI system end to end?”
- “How would you design and implement an LLM application?”
When would you choose a fixed workflow, a single agent, or a multi-agent system?
Asked in 3 of 942 interviews · last asked Aug 18 · Concept · Senior
Other ways it was asked
- “Why might you build a fixed workflow when agent technology is available?”
- “When would you use a fixed workflow, a single agent, or a multi-agent system?”
- “Why did you use the multi-agent architecture instead of just using a single LLM call?”
How would you reduce latency in a voice-based AI platform?
Asked in 2 of 942 interviews · last asked Sep 24 · Scenario · Senior
Other ways it was asked
- “How would you reduce latency in an AI system?”
What steps would you take to take a generative AI project from scratch to production?
Asked in 2 of 942 interviews · last asked Jul 14 · Design · Senior
Other ways it was asked
- “How would you take a generative AI application from initial design through production deployment?”
A long-term care pharmacy software company has about 200 facilities using its software and handles around 400 support tickets a day, with 40% covering repeated questions. Its agents have trouble finding answers in the documentation. Can AI help, and how would you approach this?
Asked in 2 of 942 interviews · last asked May 13 · Design · Junior
Other ways it was asked
- “How would you design an AI system to help a long-term-care pharmacy support team find answers in its documentation faster, serving about 200 facilities and handling 400 tickets a day, with roughly 40% repetitive tickets?”
How would you design the high-level architecture of an AI agent?
Asked in 2 of 942 interviews · last asked Jul 16 · Design · Mid-level
Other ways it was asked
- “How would you design an AI agent, including its system architecture and tool choices?”
How would you design the architecture of an AI system?
Asked in 2 of 942 interviews · last asked Jul 16 · Design · Senior
Other ways it was asked
- “Tell me about an AI system you built and explain its architecture.”
- “Describe an AI solution you built: what model did you choose, why did you choose it, and what was the architecture?”
How would you build an MCP server, and what components would it include?
Asked in 2 of 942 interviews · last asked Sep 8 · Design · Mid-level
Other ways it was asked
- “How would you architect an application using an MCP server, including the client-side components?”
- “How would you design an MCP server for an application, and what three tools or capabilities would it expose?”
How would you design a production multi-agent system end to end?
Asked in 2 of 942 interviews · last asked Oct 7 · Design · Senior
Other ways it was asked
- “How would you orchestrate a multi-agent system?”
Can AI systems be built using open-source or closed-source models, and how do those approaches differ?
Asked in 2 of 942 interviews · last asked Oct 13 · Concept · Junior
Other ways it was asked
- “What is the difference between open-source and closed-source AI models?”
How would you design an AI system to determine whether a site unit is ready for site acceptance testing, identify blockers, and provide the right contact?
Asked once · last asked Sep 29 · Design · Senior
Other ways it was asked
- “How would you design an AI system to assess a site unit's readiness for site acceptance testing?”
- “How would you design a real-time AI system to determine whether a site unit is ready for site acceptance testing, identify blockers, and provide the right contact?”
- “How would you build an AI system for a field engineer to check whether a site unit is ready for site acceptance testing?”
What components would an agent-based system need to convert test scenarios into Playwright tests?
Asked once · last asked Oct 8 · Design · Senior
Other ways it was asked
- “How would you build an agent to convert test scenarios into Playwright tests?”
- “How long might it take to build an agent that converts test scenarios into Playwright tests?”
How would you design an AI system to enhance customer support?
Asked once · last asked Jun 22 · Design · Mid-level
Other ways it was asked
- “Which technique would you use to build an AI system that enhances customer support?”
How would you design the tools and data pipeline for an AI scorecard agent?
Asked once · last asked Jul 25 · Design · Senior
Other ways it was asked
- “What model and overall architecture would you use for an AI scorecard agent?”
How would you implement a chatbot if the RAG pipeline takes a long time to respond?
Asked once · last asked Sep 10 · Scenario · Senior
Other ways it was asked
- “How would you implement asynchronous processing for a chatbot with a slow RAG pipeline beyond simply using an async function or queue?”
How would you design and implement an internal low-code/no-code generative AI application to accelerate functional scoping for projects?
Asked once · last asked Sep 20 · Design · Senior
What optimizations would you make to reduce tool calls or scale the tool execution pipeline?
Asked once · last asked Sep 22 · Design · Senior
Other ways it was asked
- “What optimizations would you implement to reduce tool call costs?”
If the university currently uses Microsoft 365 for their productivity suite, how would that factor into your implementation plan for Gemini and Google Workspace?
Asked once · last asked Sep 24 · Scenario · Senior
Other ways it was asked
- “If a university is set on continuing to use Microsoft 365 (Word, Outlook, Calendar, etc.), how would that change or factor into your implementation plan for Gemini and Google Workspace adoption?”
How would you design an AI bot that converts natural-language questions into safe SQL queries against a customer database?
Asked once · last asked Sep 25 · Design · Senior
Other ways it was asked
- “How would you design an AI bot that accepts natural-language questions, queries a customer's SQL database, and responds through a website?”
If the application has direct database access, why use an LLM, and how would you design the solution?
Asked once · last asked Sep 25 · Design · Senior
Other ways it was asked
- “If we have direct database access, why use an LLM to query the database, and why make the design complicated?”
How would you build an assistant that provides accurate, up-to-date, cited, access-controlled answers from multiple data sources?
Asked once · last asked Sep 29 · Design · Senior
Other ways it was asked
- “How would you build an assistant that provides accurate, up-to-date, cited, access-controlled answers across more than 100 tools and data sources?”
How would you connect an external website to a chatbot and provide the website's data to it?
Asked once · last asked Sep 30 · Design · Mid-level
Other ways it was asked
- “How would you integrate an external company website with a chatbot so the bot can use the website's data?”
If you want to set up a new multi-agent assistant system for question answering and task execution, what tools would you use across the toolchain?
Asked once · last asked Oct 2 · Design · Senior
Other ways it was asked
- “What tools would you use across the toolchain to build a multi-agent assistant system?”
How would you design a fabric quality-control system that detects stains, stitching issues, holes, and color inconsistencies, and prioritizes items for human review?
Asked once · last asked Oct 21 · Design · Mid-level
How would you build a RAG system to answer customer queries using both PDF documents and data from a database? Walk me through the steps and components.
Asked once · last asked Feb 5 · Design · Senior
For speech-to-text and text-to-speech integrations with a PBX, when would you process audio in real time versus in batches?
Asked once · last asked Feb 19 · Concept · Mid-level
What role can a RAG pipeline play in a voice-based AI platform, and why might you use one?
Asked once · last asked Mar 12 · Design · Senior
How would you design a RAG system, including the business problem it solves and its architecture?
Asked once · last asked Mar 16 · Design · Senior
How would you design a real-time anomaly detection platform that processes video from hundreds of cameras, identifies suspicious activities using AI models, and alerts human reviewers in under two seconds, addressing architecture, major components, scalability, low latency, and reliable alert delivery?
Asked once · last asked Mar 27 · Design · Senior
Other ways it was asked
- “Design a real-time anomaly detection platform that processes video streams from hundreds of cameras, identifies suspicious activities using AI models, and alerts human reviewers in under two seconds. Explain the system architecture, its major components, and how you would ensure scalability, low latency, and reliable alert delivery.”
How would you build an intelligent recommendation agent, and what technologies would you use?
Asked once · last asked Apr 10 · Design · Mid-level
How would you design a maintainable, evaluable LLM system?
Asked once · last asked Apr 10 · Design · Mid-level
Could you explain how to build scalable, production-ready LLM systems?
Asked once · last asked Apr 10 · Design · Senior
How would you approach building a collision-avoidance system using AI and machine-learning techniques?
Asked once · last asked Apr 15 · Design · Senior
How would you quickly develop and scale an AI/ML-based module to assess covariance or uncertainty for a catalog of about 20,000 objects?
Asked once · last asked Apr 15 · Design · Senior
How would you approach implementing an AI workflow or ETL process in the cloud, and how would you address the main challenges?
Asked once · last asked Apr 17 · Behavioral · Mid-level
Other ways it was asked
- “Could you describe a specific project in which you implemented an AI workflow or ETL process in the cloud, including the challenge and how you approached it?”
If you were designing an enterprise-grade agentic system from scratch today, what major architectural decisions would you need to make? Talk me through how you would approach the problem and the aspects you would cover.
Asked once · last asked Apr 17 · Design · Senior
What design patterns would you use to scale an AI agent to 10,000 users?
Asked once · last asked Apr 17 · Design · Senior
How would you integrate or deploy a Granite model with Watson AI on AWS GovCloud, and what architecture would you choose among AWS-native services, inference containers, or APIs?
Asked once · last asked Apr 20 · Design · Senior
Other ways it was asked
- “How did you integrate or deploy the Granite model with Watson AI on AWS GovCloud? What was the integration architecture, and did you use AWS-native services, inference containers, or APIs to run it?”
How would you securely extract and evaluate information from a large PDF corpus for AI agents, comparing AWS or GCP tools, Palantir Foundry, and private cloud to right-size the architecture for a customer?
Asked once · last asked Apr 20 · Design · Senior
Other ways it was asked
- “How would you securely and accurately extract information from a large PDF corpus, evaluate the results, and make the information available to AI agents? How would you compare AWS or GCP tools, Palantir Foundry, and private cloud when right-sizing the architecture for a customer?”
If a customer needs a RAG system for about 1,000–2,000 users, with 100 concurrent users, how would you determine the LLM's tokens-per-minute configuration?
Asked once · last asked Apr 24 · Scenario · Senior
How would you migrate or rebuild an AI-driven marketing initiative using a modern AI-first stack, such as the Anthropic API, OpenAI, or a modern CDP, and why would you make those changes?
Asked once · last asked Apr 26 · Design · Senior
Other ways it was asked
- “How would you migrate or rebuild an AI-driven marketing initiative in a modern AI-first stack using an API such as Anthropic or OpenAI, or a modern CDP? What would you change, and why?”
How would you scale the churn-prediction proof of concept to all 70 million users in the next 60 days without breaking it, and what is the one thing you’re most worried will fail?
Asked once · last asked Apr 26 · Design · Senior
What challenges can arise when implementing an AI workflow end to end, and how would you address them?
Asked once · last asked Apr 27 · Scenario · Senior
In LLM applications, how do you manage context and memory over a long interaction when the full context cannot fit in the model's context window?
Asked once · last asked Apr 28 · Design · Senior
How do you maintain conversation state in an AI agent application?
Asked once · last asked Apr 29 · Scenario · Mid-level
What factors do you consider when evaluating AI solutions, including agentic AI, AI governance, cloud workloads, and API integrations?
Asked once · last asked May 5 · Concept · Senior
How do AI workloads differ from traditional workloads in their data center power, space, and cooling requirements, and what drives increasing demand for these resources?
Asked once · last asked May 7 · Concept · Senior
How would you design a document-processing workflow that extracts and embeds a user's document, then determines eligibility based on its contents?
Asked once · last asked May 12 · Design · Mid-level
How would you approach building a system for a moving and storage company that reads customer emails and extracts the pickup and drop-off addresses, move date, number of rooms, and special items such as pianos or fragile belongings?
Asked once · last asked May 13 · Design · Mid-level
How would you estimate the load on a RAG system serving 185 universities?
Asked once · last asked May 13 · Design · Mid-level
What scaling challenges can arise when deploying a RAG system across 185 universities?
Asked once · last asked May 13 · Design · Mid-level
How do you design and build Model Context Protocol (MCP) services?
Asked once · last asked May 19 · Design · Senior
Can you describe the technical architecture of an AI chatbot and explain how you would design and implement it?
Asked once · last asked May 22 · Design · Senior
How would you scale an AI chatbot to serve millions of users concurrently and ensure it remains reliable?
Asked once · last asked May 22 · Design · Senior
How does the backend of a RAG chatbot work, and how would you design its implementation?
Asked once · last asked May 22 · Design · Senior
How would you integrate an AI chatbot with a booking system?
Asked once · last asked May 22 · Design · Mid-level
How would you design an agentic AI workflow with LangChain, CrewAI, or a similar framework, covering its use case, orchestration, tool usage, memory, and reliability?
Asked once · last asked May 27 · Design · Senior
Other ways it was asked
- “Describe an agentic AI workflow you built with LangChain, CrewAI, or a similar framework, including its use case, orchestration, tool usage, memory, and reliability.”
How would you design an end-to-end RAG pipeline, including its data sources and architectural or machine-learning choices?
Asked once · last asked Jun 12 · Design · Senior
How would you design a RAG system to answer questions about a company's historical revenue and profit?
Asked once · last asked Jun 12 · Scenario · Mid-level
How do you scale systems to millions of users while addressing security, infrastructure, and the components of an AI solution beyond its models and agents?
Asked once · last asked Jun 12 · Design · Senior
When integrating open-source projects into an AI coding assistant, how do you ensure scalability and performance?
Asked once · last asked Jun 17 · Design · Senior
How would you design an AI agent that processes high-level design documents and generates detailed designs and test cases, beyond relying on prompts and templates?
Asked once · last asked Jun 25 · Design · Senior
How would you build a hybrid search system that combines vector retrieval with traditional keyword search?
Asked once · last asked Jul 16 · Design · Senior
How would an AI sales agent trained on a retailer's products work?
Asked once · last asked Jul 16 · Scenario · Mid-level
How would you design the end-to-end architecture for an AI sales workflow in a retail store?
Asked once · last asked Jul 16 · Design · Mid-level
What challenges does an AI sales agent face that a standard chatbot does not?
Asked once · last asked Jul 16 · Concept · Mid-level
Why is async programming important in a GenAI system?
Asked once · last asked Jul 16 · Concept · Mid-level
What considerations would you evaluate when choosing between cloud-based machine learning infrastructure and an entirely on-device approach?
Asked once · last asked Jul 1 · Design · Senior
If you were implementing a generative AI solution for personalized marketing content at scale, what key architectural components would you design, and how would you address data privacy, model bias, and content diversity?
Asked once · last asked Jul 16 · Design · Senior
If you had to build an agentic AI system, what would be your plan? How would you design this system?
Asked once · last asked Jul 16 · Design · Senior
How would you integrate an AI model through an API into an application?
Asked once · last asked Jul 16 · Scenario · Mid-level
How would you store short-term, medium-term, and long-term memory in a multi-turn chatbot?
Asked once · last asked Jul 16 · Design · Senior
Walk me through the expected flow of an AI agent calling an insurer to report a car accident.
Asked once · last asked Jul 16 · Design · Senior
What should happen when an AI agent calls the insurer's customer service number on a car insurance card?
Asked once · last asked Jul 16 · Scenario · Mid-level
How should an AI phone agent handle a call that goes to voicemail?
Asked once · last asked Jul 16 · Scenario · Mid-level
If an insurer does not provide a callback after an AI agent reaches voicemail, how should the system follow up?
Asked once · last asked Jul 16 · Scenario · Senior
How would you integrate an LLM API while accounting for errors, cost, latency, and moderation?
Asked once · last asked Jul 21 · Design · Senior
How would you integrate an LLM into an application and handle errors?
Asked once · last asked Jul 22 · Scenario · Mid-level
How would you talk to a founding engineer about building their entire AI infrastructure on Cloudflare?
Asked once · last asked Jul 24 · Scenario · Senior
How would you architect the high-level tech stack and data flow for a chatbot application?
Asked once · last asked Jul 24 · Design · Mid-level
If you had to design a ChatGPT-like system that reads PDF documents, what would your design approach be?
Asked once · last asked Jul 25 · Design · Senior
How would you design a multi-agent system with one agent retrieving data, one storing and managing data, and one refining the final response?
Asked once · last asked Jul 25 · Design · Senior
Other ways it was asked
- “If you need to design a multi-agent system with one agent retrieving data, one storing/managing data, and one fine-tuning the final response, what would your design approach be?”
How would you integrate an LLM into a chatbot?
Asked once · last asked Jul 25 · Scenario · Mid-level
Can you explain the architecture of an AI-powered calling and SMS system, including the caller experience and which components make each decision?
Asked once · last asked Jul 26 · Design · Senior
Why might deterministic business rules belong in application code rather than an AI prompt?
Asked once · last asked Jul 26 · Concept · Senior
What could go wrong if you changed an AI prompt but not the worker code that enforces the same rule?
Asked once · last asked Jul 26 · Scenario · Senior
How would you design and implement a multilingual voice AI agent with agentic tool orchestration?
Asked once · last asked Jul 27 · Design · Senior
How do you think datasets containing millions of data points from government buyers and sellers could be used to create a chatbot that answers their queries and provides a first point of contact for issue resolution?
Asked once · last asked Jul 27 · Scenario · Mid-level
In a Retell integration, what does the worker control, and what does Retell control?
Asked once · last asked Jul 27 · Concept · Mid-level
How does a sentiment classification pipeline work end to end, from input through automated decisions to downstream actions without human review?
Asked once · last asked Jul 27 · Design · Senior
How do you prioritize which AI initiatives to start with, and how do you make sure you're building a cohesive end-to-end solution rather than a collection of point solutions?
Asked once · last asked Jul 27 · Scenario
How would you design policy layers for an AI agent at the channel, product, and system levels, accounting for social media inputs, employee communications, and file structures?
Asked once · last asked Jul 27 · Design · Senior
How would you determine a user's current stage in a chat conversation?
Asked once · last asked Jul 27 · Scenario · Mid-level
From an infrastructure perspective, how would you design a GPU cluster and select the right GPUs for model customization?
Asked once · last asked Jul 27 · Design · Senior
How would you build a tool-using chat agent like Claude, and which libraries could speed up development?
Asked once · last asked Jul 28 · Design · Senior
How would you design the backend for a chatbot, including its core entities, chat flow, streaming, and context management?
Asked once · last asked Jul 28 · Design · Senior
How would you approach building a system where an agent gets the customer's name from a CRM by phone number before calling?
Asked once · last asked Jul 28 · Design · Mid-level
How would you assess data privacy requirements and choose a deployment approach for an agentic workflow?
Asked once · last asked Jul 28 · Scenario · Senior
How would you implement speech-to-text and text-to-speech in a chatbot?
Asked once · last asked Jul 29 · Design · Mid-level
Other ways it was asked
- “If I want to build a chatbot from scratch and add speech-to-text, what would you do?”
What components would you include in an MCP-based AI QA agent framework built from scratch?
Asked once · last asked Jul 30 · Design · Senior
How would you build or optimize an LLM-powered or agentic application, including prompt design, tool use, and orchestration?
Asked once · last asked Aug 2 · Design · Senior
Other ways it was asked
- “How have you approached building or optimizing LLM-powered or agentic applications? Walk me through one example, including how you handled prompt design, tool use, or orchestration.”
How would you architect a complex agentic or LLM system end to end, and address its hardest engineering challenges?
Asked once · last asked Aug 2 · Design · Senior
Other ways it was asked
- “What is the most complex agentic or LLM system you have built end to end? Walk me through its architecture and the hardest engineering problem you had to solve.”
How do AI agents and DAG-based systems work, and when would you use each?
Asked once · last asked Aug 4 · Concept · Mid-level
What advantages does an MCP server offer over a conventional server or regular backend logic?
Asked once · last asked Aug 4 · Concept · Mid-level
How would you prevent an LLM from treating a US Discovery ad as proof for the German market if it can ignore or overwrite an “international—not proof” metadata tag?
Asked once · last asked Aug 4 · Design · Senior
In an AI marketing workflow, what should the analysis step do, and what output should it produce before a compliance gate?
Asked once · last asked Aug 5 · Design · Senior
How would you design an end-to-end AI marketing solution rather than addressing only one bottleneck?
Asked once · last asked Aug 5 · Design · Senior
How would you enforce a consistent schema across the steps in an AI workflow?
Asked once · last asked Aug 5 · Design · Senior
How would you handle a failed compliance gate so the workflow can fail safely and recover gracefully?
Asked once · last asked Aug 5 · Scenario · Senior
Would you include a feedback loop or safety mechanism, such as an LLM-as-a-judge pattern, in an AI marketing workflow?
Asked once · last asked Aug 5 · Design · Senior
For an AI marketing workflow, should an intake form use free-text prompts or structured fields, and what information should it collect?
Asked once · last asked Aug 5 · Design · Mid-level
How would you apply AI automation to an IoT integration?
Asked once · last asked Aug 6 · Scenario · Mid-level
What architectural components are needed to make an AI agent system production-ready?
Asked once · last asked Aug 7 · Design · Senior
Other ways it was asked
- “What architectural components would be missing from an AI agent system before it is ready for production?”
How would you design the architecture for a ticket-handling AI agent?
Asked once · last asked Aug 7 · Design · Senior
Other ways it was asked
- “What component flow would you use for a ticket-handling AI agent system?”
If the access-granting system is a black box, would the ticket-handling agent still need its own database?
Asked once · last asked Aug 7 · Design · Senior
How would you provide an LLM with information about relationships within a large dataset without overwhelming its context window?
Asked once · last asked Aug 7 · Design · Senior
Other ways it was asked
- “How would you give an LLM the interconnectedness of a large dataset without overwhelming it?”
How can LLMs be used to solve technically challenging problems in software products?
Asked once · last asked Aug 7 · Scenario · Senior
How would you design a voice-recognition system that lets a user say, “I want to retrieve all contracts”?
Asked once · last asked Aug 11 · Design · Mid-level
How would you design a layered system for speech recognition, processing, and delivering responses through Telegram or email, and what bottlenecks would you anticipate?
Asked once · last asked Aug 11 · Design · Mid-level
How would you prevent the same task from being processed more than once when speech transcripts differ slightly?
Asked once · last asked Aug 11 · Scenario · Mid-level
At what stage would you check for duplicate or previously asked questions when the system must respond immediately?
Asked once · last asked Aug 11 · Design · Mid-level
At what stage should a system check whether a question or answer has appeared before?
Asked once · last asked Aug 11 · Design · Mid-level
Why might you analyze transcribed text before sending it through the rest of the processing pipeline?
Asked once · last asked Aug 11 · Design · Mid-level
How would you scale a system that must analyze large volumes of transcripts quickly when each transcript may lack sufficient context?
Asked once · last asked Aug 11 · Design · Senior
Walk me through how you would lower a PyTorch model from a high-level graph into hardware-specific code.
Asked once · last asked Aug 12 · Design · Senior
How would you automate an AI-powered process while managing token limits and Gemini API keys?
Asked once · last asked Aug 14 · Scenario · Mid-level
How would you design a seven-agent architecture, and what purpose would each node serve?
Asked once · last asked Aug 18 · Design · Mid-level
How do you define the boundary between agents when there are 150 applications? Does each agent touch all the data, or do you segment them by system or function?
Asked once · last asked Aug 19 · Design · Senior
Given the mixed landscape of COBOL and newer applications, how would you architect and govern agentic AI adoption across that environment?
Asked once · last asked Aug 19 · Design · Senior
How would you integrate agentic SDLC automation tools into modernization programs that combine legacy and newer technology stacks?
Asked once · last asked Aug 19 · Design · Senior
How would you design a RAG chatbot that handles both structured data and unstructured data with hierarchical entity relationships?
Asked once · last asked Aug 20 · Design · Senior
Beyond state management, what high-level architectural components would you define for an agent orchestration harness?
Asked once · last asked Aug 20 · Design · Senior
How would you guide an executive team through the complexities of architecting a scalable GenAI application, specifically navigating the trade-offs between managed APIs, complex RAG pipelines, and fine-tuning open-weight models in production?
Asked once · last asked Aug 20 · Design · Senior
How would you design a RAG system end to end, reduce hallucinations, and evaluate its performance?
Asked once · last asked Aug 22 · Design · Senior
How would you manage model selection, cost, latency, hallucinations, and data privacy in a GenAI application?
Asked once · last asked Aug 22 · Scenario · Senior
How are internal assistant agents built in practice, and what works versus quietly fails?
Asked once · last asked Aug 22 · Design · Senior
What are the trade-offs between one shared internal AI assistant and multiple assistants tailored to different teams?
Asked once · last asked Aug 22 · Design · Senior
What factors would influence your choice between one shared AI assistant and multiple team-specific assistants?
Asked once · last asked Aug 22 · Design · Senior
Other ways it was asked
- “What factors influenced the choice between one shared assistant and multiple team-specific assistants?”
Could you describe the system interface or harness that wraps around the model?
Asked once · last asked Aug 22 · Design · Senior
Can you tell me more about how model routing works within the system? Does it involve using several models for different tasks?
Asked once · last asked Aug 22 · Design · Senior
How would you design an AI assistant's memory of past conversations, including what information to retain or persist across sessions?
Asked once · last asked Aug 22 · Design · Mid-level
Other ways it was asked
- “Can you explain whether the assistant can remember past conversations? If so, what information is retained or persisted across sessions?”
How would you choose whether an AI agent runs in a user's terminal, a browser, a server, or elsewhere, and enable it to perform tasks autonomously without active supervision?
Asked once · last asked Aug 22 · Design · Senior
Other ways it was asked
- “Where does the system typically run: in a user's terminal, a browser, a server, or elsewhere? Can it perform tasks autonomously when no one is actively watching?”
How would you integrate an AI agent with external tools and systems using APIs, servers, or other methods, and assign responsibility for those integrations?
Asked once · last asked Aug 22 · Design · Senior
Other ways it was asked
- “How does it reach or integrate with external tools and systems? Does it use APIs, servers, or some other method, and who is responsible for these integrations?”
How would you decide how many tools to integrate into an AI agent system and which to expose to the agent, considering the effects on performance and complexity?
Asked once · last asked Aug 22 · Design · Senior
Other ways it was asked
- “How many tools are integrated within the system? Does having many tools affect performance or complexity, and how do you determine which tools to expose to the agent?”
When you build an LLM-based feature, what are the first three things you decide upfront—such as model choice, prompting approach, and RAG or tools—and why?
Asked once · last asked Aug 24 · Design · Senior
How would you implement state management, idempotency, and recovery or compensation in a multi-step agent workflow where steps can fail?
Asked once · last asked Aug 24 · Behavioral · Senior
Other ways it was asked
- “Describe a multi-step agent workflow you built where steps can fail. How did you implement state management, idempotency, and recovery or compensation?”
How would you implement AI features in a learning management system, such as feedback, hints, and worksheet generation?
Asked once · last asked Aug 25 · Scenario · Mid-level
When comparing OpenAI models for an AI chatbot, how do you account for concurrency, and what factors guide your choice?
Asked once · last asked Aug 25 · Scenario · Senior
When two users use an AI chatbot at the same time, how are token limits and usage allocated?
Asked once · last asked Aug 25 · Concept · Mid-level
How would you design an AI integration and an API processing architecture for one million incoming records?
Asked once · last asked Aug 25 · Design · Senior
Can you walk through how an agentic AI restaurant search system would actually work?
Asked once · last asked Aug 25 · Design · Mid-level
How would you structure the observation-action loop in an autonomous browser to handle dynamic page elements, dynamic loading, and shadow DOMs?
Asked once · last asked Aug 25 · Behavioral · Senior
Other ways it was asked
- “How did you structure the observation-action loop to handle dynamic page elements, dynamic loading, and shadow DOMs in an autonomous browser project?”
How would you define an end-to-end latency budget for an ML model pipeline?
Asked once · last asked Aug 25 · Design · Senior
How would you manage memory in multi-tenant agentic workflows?
Asked once · last asked Aug 25 · Design · Senior
How would you design short-term and working memory for an autonomous agent?
Asked once · last asked Aug 25 · Design · Senior
How would you design an AI agent with read/write access to data or code to prevent cascading hallucinations and infinite loops?
Asked once · last asked Aug 26 · Design · Senior
How would you decompose an ambiguous, complex workflow into sequential deterministic subtasks for autonomous agents, execute them reliably, and integrate the results?
Asked once · last asked Aug 26 · Design · Senior
Can you describe a specific example of decomposing a complex problem into sequential subtasks for AI agents?
Asked once · last asked Aug 26 · Design · Senior
How would you choose models for each stage of a pipeline while managing cost, latency, and output reliability?
Asked once · last asked Aug 27 · Design · Senior
How would you build an internal application using offline LLMs to prevent company data from being exposed?
Asked once · last asked Aug 28 · Design · Mid-level
How would you design an end-to-end system for fake news detection and credibility assessment?
Asked once · last asked Aug 28 · Design · Mid-level
What are your views on the AI infrastructure landscape and where it is headed?
Asked once · last asked Aug 28 · Senior
If detection runs in the cloud rather than on the edge device, how can that help filter false positives?
Asked once · last asked Aug 28 · Scenario · Mid-level
How would you orchestrate an LLM pipeline for classification, and when would you use one call versus multiple calls?
Asked once · last asked Sep 2 · Design · Senior
Other ways it was asked
- “What other calls might an LLM-based classification pipeline make?”
When would you use a sequence of LLM calls versus one agent with multiple roles that iterates over user input?
Asked once · last asked Sep 2 · Design · Senior
Would you use the same LLM for every task in a pipeline or select different models for different tasks?
Asked once · last asked Sep 2 · Design · Senior
How would you build a PDF-to-chat application, from processing multiple PDFs through storing the data and generating chat responses?
Asked once · last asked Sep 4 · Design · Senior
How would you build an AI chatbot for college students that handles tasks and practice recommendations?
Asked once · last asked Sep 5 · Design · Mid-level
How would you build a chatbot that answers questions using 10,000 documents?
Asked once · last asked Sep 5 · Design · Senior
If an AI model classifies chest pain as level one while a hard-coded override assigns level three, how should the application resolve the conflict?
Asked once · last asked Sep 6 · Scenario · Mid-level
How would you build a person detection and tracking solution that triggers an alarm when someone stays in or crosses a monitored area?
Asked once · last asked Sep 7 · Design · Senior
How would you change the architecture to reduce LLM costs for a chatbot handling 14,000 incident resolution requests per day?
Asked once · last asked Sep 8 · Design · Senior
How can resolved LLM query data be used to reduce future LLM costs and handle recurring questions internally?
Asked once · last asked Sep 8 · Design · Senior
How would you design an AI agent for a specific business task, build its pipeline, and ensure it operates reliably?
Asked once · last asked Sep 8 · Design · Mid-level
Where would you enforce maximum node retries, loop detection, timeouts, and token or cost budgets in an agent workflow?
Asked once · last asked Sep 8 · Design · Senior
How would you implement idempotency and state consistency for write tools used by an AI agent?
Asked once · last asked Sep 8 · Design · Senior
How would you design a production LLM feature using AWS Bedrock?
Asked once · last asked Sep 9 · Design · Senior
How would you turn an AI pipeline into an end-to-end product such as a chatbot?
Asked once · last asked Sep 10 · Design · Senior
How would you design a chatbot that uses RAG and can also trigger other actions?
Asked once · last asked Sep 10 · Design · Senior
What happens in a chatbot's request flow after an analyst submits a query?
Asked once · last asked Sep 10 · Design · Mid-level
How would you manage conversation history, memory, and context windows in a chatbot?
Asked once · last asked Sep 10 · Design · Senior
How would you handle a chatbot conversation that exceeds the model's context window?
Asked once · last asked Sep 10 · Scenario · Senior
What are the main technical challenges in building a lightweight model that runs locally and syncs with the cloud when connectivity returns?
Asked once · last asked Sep 10 · Design · Mid-level
How would you design an AI-based college attendance system?
Asked once · last asked Sep 10 · Design · Mid-level
What are your initial thoughts and overall approach to building the patient risk prediction platform?
Asked once · last asked Sep 11 · Design · Senior
If a camera produces 30 FPS but your model processes only 10 FPS, what could you do to handle the throughput gap?
Asked once · last asked Sep 11 · Scenario · Senior
How would you design an end-to-end semantic search system, from user query to ranked results, and measure its improvement?
Asked once · last asked Sep 16 · Design · Senior
Can you walk me through the end-to-end architecture of a real-time multimodal voice AI agent from microphone input to spoken response, including latency budgets and session degradation handling?
Asked once · last asked Sep 16 · Design · Senior
How would you ensure an AWS-hosted machine learning solution remains scalable as data volume or traffic increases?
Asked once · last asked Sep 17 · Design · Senior
How would you mitigate risks when pursuing AI-driven recommendations or expanding product and logistics capabilities, given a client's legacy technology?
Asked once · last asked Sep 17 · Scenario · Senior
How would you approach building a scalable NLP application for sentiment analysis? What particular tools or frameworks would you consider essential, and why?
Asked once · last asked Sep 17 · Design · Senior
Other ways it was asked
- “What challenges might arise when deploying a scalable NLP sentiment analysis system, and how would you address them?”
How would you begin integrating an LLM into a live chat where latency and cost are critical?
Asked once · last asked Sep 18 · Design · Senior
How would you design real-time voice input for a chat, with an experience similar to a phone call?
Asked once · last asked Sep 18 · Design · Senior
Other ways it was asked
- “How would you approach the high-level design for adding real-time voice to an existing AI chat?”
How would you design a production AI chat application with React and ASP.NET Core to stream responses while controlling latency and cost?
Asked once · last asked Sep 18 · Design · Senior
How would you store and trim conversation context when a chat history has grown to 300 messages?
Asked once · last asked Sep 18 · Design · Senior
How would you use embeddings and RAG in production to find relevant older messages in a chat history?
Asked once · last asked Sep 18 · Design · Senior
How would you design intelligent task assignment for an employee-management application to improve process efficiency?
Asked once · last asked Sep 21 · Design · Senior
Other ways it was asked
- “How would you design intelligent task assignment for an employee-management application to improve process efficiency by 40%?”
How can AI or AWS services be used to automate task assignment in an employee-management application?
Asked once · last asked Sep 21 · Design · Senior
How would you choose and implement an approach for a given use case: traditional ML, deep learning (such as an autoencoder, neural network, or CNN), an agentic system, RAG, or fine-tuning?
Asked once · last asked Sep 22 · Design · Mid-level
Other ways it was asked
- “How would you determine, based on a problem's use case, whether to use a traditional ML solution, a deep learning model (such as an autoencoder, neural network, or CNN), an agentic system, RAG, or fine-tuning, and how would you implement that choice?”
Page 1 · Page 2
More System Design interview questions
- All System Design interview questions
- System Design Fundamentals Interview Questions
- Software Architecture Interview Questions
- Microservices and Integration Interview Questions
- API Design and Messaging Interview Questions
- Database and Data Pipeline System Design Interview Questions
- Scalability and Reliability Interview Questions
- Cloud Architecture and Multitenancy Interview Questions
- System Design Interview Case Study Questions
How this list is made
Every question here was asked by a real interviewer in a real job interview that a candidate ran MeetAssist in. After the call, the interviewer's questions are pulled from the transcript. Practice sessions, tests, exams and anything that was not a job interview are dropped.
Questions are rewritten so they make sense on their own: names of people, employers and clients are removed, and speech-to-text mistakes are fixed. Small talk, logistics and questions about one candidate's own résumé are left out. Rephrasings of the same question are merged, and the count is the number of different interviews it came up in.
The interviews come from candidates who use MeetAssist, so they lean towards software, data and AI roles: the shares describe these interviews, not every System Design interview. Answers to the most asked questions are written and checked by an editor; a "common mistake" is shown only when reviews of real answers in these interviews show it, with the count.
942 interviews, 495 candidates, Oct 2025 to Oct 2026. Updated every month with new interviews.
Or add the Chrome extension