ML System Design interview questions

Showing 181–210 of 332 results

• Design image and multimodal generation systemsHard

  This question evaluates a candidate's competence in designing end-to-end image and multimodal generation systems, covering data collection and curation, model architecture and conditioning choices, training objectives, safety and content filtering, evaluation metrics, deployment, monitoring, and critical analysis of relevant research.

  MetaMachine Learning EngineerML System DesignTechnical Screen
• Design nearby place recommendationsHard

  This question evaluates skills in designing real-time, large-scale machine learning recommender systems that combine geospatial candidate generation, feature engineering, ranking models, personalization, privacy controls, fraud detection, and operational constraints like latency and throughput.

  MetaMachine Learning EngineerML System DesignOnsite
• Design place-of-interest ML systemHard

  Design place-of-interest ML system evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  MetaSoftware EngineerML System DesignOnsite
• Design weapon-selling ad detection from postsMedium

  This question tests the ability to design an end-to-end multimodal classification system for detecting policy-violating content (weapon sales) while balancing latency, precision, and robustness against adversarial evasion.

  MetaMachine Learning EngineerML System DesignOnsite
• Explain an AI System You Designed and the Problem It SolvedMedium

  Tests your ability to articulate an end-to-end production AI/ML system you designed, detailing the business justification, technical architecture, evaluation strategy, failure modes, and your individual contributions.

  MetaSoftware EngineerML System DesignTechnical Screen
• Prevent Private Code Leakage in Coding AgentsMedium

  This problem tests the ability to design an ML-driven safeguard system that prevents private codebase exposure during the training, fine-tuning, and inference phases of a coding agent.

  MetaMachine Learning EngineerML System DesignTechnical Screen
• Calibrate LLM output to match Word formattingMedium

  This question tests the ability to design an end-to-end system for enforcing strict structured formatting (e.g., Docx/Markdown) on LLM-generated text, focusing on constraint enforcement and reliability.

  MicrosoftMachine Learning EngineerML System DesignOnsite
• Design Chatbot Personalization MemoryMedium

  This problem evaluates the ability to design a scalable, low-latency memory system for an LLM-based chatbot that stores and retrieves user-specific facts to provide personalized responses.

  MicrosoftSoftware EngineerML System DesignOnsite
• Design Product-Description Content Moderation at Marketplace ScaleMedium

  Design an ML moderation system for multilingual seller descriptions at marketplace scale. Cover policy-aware labels, severity-based actions, calibrated thresholds, human review and appeals, low-latency serving, selective rescans, auditability, drift, and adversarial abuse.

  MicrosoftData ScientistML System DesignTechnical Screen
• Design a Product Search SystemMedium

  This question evaluates skills in designing scalable, low-latency product search systems, covering information retrieval, ranking model design, feature generation, data ingestion and indexing pipelines, and serving infrastructure.

  MicrosoftMachine Learning EngineerML System DesignOnsite
• Design a RAG Ranking PipelineMedium

  This problem tests your ability to design a multi-stage retrieval and ranking architecture for a RAG system, focusing on balancing search accuracy, latency, and system scalability.

  MicrosoftSoftware EngineerML System DesignOnsite
• Design a RAG-based assistant serviceMedium

  This question assesses the ability to architect a scalable Retrieval-Augmented Generation (RAG) system, focusing on data indexing, retrieval strategy, and mitigating common LLM issues like hallucinations and safety.

  MicrosoftMachine Learning EngineerML System DesignOnsite
• Design a Retrieval-Grounded Local Sports News RecommenderMedium

  This question tests the ability to design an end-to-end RAG system for news, focusing on temporal data relevance, hybrid retrieval strategies, and mitigation of LLM hallucinations.

  MicrosoftMachine Learning EngineerML System DesignHR Screen
• Design email ranking and summarization in OutlookMedium

  This problem tests the ability to architect a scalable ML pipeline that balances personalized email prioritization with LLM-based summarization while addressing low-latency requirements and data safety.

  MicrosoftMachine Learning EngineerML System DesignOnsite
• Design quality checks for spreadsheet LLM dataMedium

  This problem tests the ability to design an end-to-end data validation pipeline for spreadsheet data and determine the necessity of fine-tuning a Large Language Model (LLM) for specific structured data tasks.

  MicrosoftMachine Learning EngineerML System DesignTechnical Screen
• Explain Transformers and deploy an LLM safelyEasy

  This question evaluates understanding of Transformer architectures and practical LLM deployment competencies, covering attention mechanisms, token and positional representations, computational complexity, and production concerns like latency, cost, quality, safety, and privacy.

  MicrosoftMachine Learning EngineerML System DesignTechnical Screen
• Predict Employee Attrition from an HR Dataset and Explain Its DriversMedium

  A machine learning design exercise: given a tab-separated HR dataset with features such as age, job role, monthly income and overtime, predict which employees will leave and explain which factors drive attrition. It covers feature processing, model choice, class imbalance, evaluation metrics, threshold selection and interpretability.

  MicrosoftApplied ScientistML System DesignOnsite
• Shard a Large Input Embedding Table Across DevicesHard

  Compare vocabulary and hidden-dimension embedding sharding through per-rank storage, token routing, hidden-state layout, skew, and downstream communication.

  MicrosoftSoftware EngineerML System DesignOnsite
• Shard an LM Head and Combine Distributed LogitsHard

  Shard an LM head by vocabulary or hidden width, distinguishing complete logits from partial dot products and exact distributed top-k from full-distribution sampling.

  MicrosoftSoftware EngineerML System DesignOnsite
• Design a Governed Agent for Portfolio RebalancingMedium

  Design an autonomous, LLM- or ML-driven portfolio rebalancing agent integrated with deterministic guardrails, pre-trade compliance checks, and audit logging. The design must guarantee safety, idempotency, strict separation of concerns, and full state reconstructibility.

  MillenniumData ScientistML System DesignTechnical Screen
• Build a Small Agent or RAG Tool with the Mistral APIMedium

  Design an end-to-end Retrieval-Augmented Generation (RAG) agent using the Mistral API to answer technical queries over internal developer documentation. The question evaluates chunking strategies, dense retrieval, prompt orchestration, latency optimizations, and evaluation methodologies.

  Mistral AIMachine Learning EngineerML System DesignTechnical Screen
• Build and design a Mistral RAG agentHard

  This question evaluates a candidate's skills in building a retrieval-augmented generation (RAG) system, covering document ingestion, chunking, embedding generation, vector indexing, retrieval, LLM API-driven streaming completions, short-term conversation memory, and operational reliability features.

  Mistral AISoftware EngineerML System DesignTechnical Screen
• Design a PDF-to-Markdown Inference APIHard

  This question evaluates system-design and ML inference orchestration skills, assessing competency in composing multi-stage pipelines that manage heterogeneous resource profiles (CPU, GPU, memory) and trade-offs between latency and throughput.

  Mistral AISoftware EngineerML System DesignTechnical Screen
• Describe model-to-GPU execution pipelineMedium

  Describe model-to-GPU execution pipeline evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  NVIDIASoftware EngineerML System DesignTechnical Screen
• Design and benchmark optimized inference pipelinesMedium

  Design and benchmark optimized inference pipelines evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  NVIDIASoftware EngineerML System DesignTechnical Screen
• Design real-time fraud detection under 50msEasy

  This question evaluates expertise in designing low-latency, high-throughput ML-powered systems for online decisioning, covering competencies in real-time inference, feature pipelines, model serving, scaling, and operational reliability within the ML System Design domain.

  NVIDIASoftware EngineerML System DesignTechnical Screen
• Discuss Transformer LLM DesignHard

  Discuss Transformer LLM Design evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  NVIDIAMachine Learning EngineerML System DesignTechnical Screen
• Explain ML compilation optimizations and hardware fitMedium

  This question evaluates a candidate's understanding of ML compiler optimizations and hardware-aware runtime strategies, assessing competencies in techniques such as kernel fusion, quantization, memory planning, scheduling/tiling, layout selection, sparsity, mixed precision, graph-level rewrites and runtime tactics.

  NVIDIASoftware EngineerML System DesignTechnical Screen
• Explain ML framework trendsHard

  This interview question evaluates ML product requirements, data/labeling, modeling, serving architecture, evaluation, monitoring, and trade-offs in a realistic interview setting. A strong answer for Explain ML framework trends states assumptions, handles edge cases, explains trade-offs, and shows how to validate the result clearly.

  NVIDIASoftware EngineerML System DesignTechnical Screen
• How would you optimize large-scale training/inference?Medium

  This problem tests your understanding of performance bottlenecks in large-scale machine learning systems, requiring you to prioritize optimization strategies across memory, compute, and communication dimensions.

  NVIDIASoftware EngineerML System DesignTechnical Screen