Practice series

AI platforms, agents & model systems

RAG, agent loops, model routing, training clusters, tool gateways, and evaluation.

18 questions27 exact tags
  1. 01

    AI Agent System

    Design a production AI agent application platform that accepts user goals, manages context and memory, calls tools safely, pauses for human review when needed,...

    Practice bank
    Preview
  2. 02

    Moderation Service

    Design a content moderation pipeline that screens user-submitted posts/media for policy violations, balancing posting latency against review accuracy at platfor...

    LinkedIn
    Preview
  3. 03

    Recommendation System

    Design a large-scale personalized recommendation system for feed/video/ads.

    Netflix / Meta
    Preview
  4. 04

    RAG Knowledge Assistant

    Design an enterprise RAG knowledge assistant that answers employee questions over internal documents (wikis, tickets, contracts) with grounded, cited answers wh...

    OpenAI
    Preview
  5. 05

    LLM Inference Gateway

    Design an LLM inference gateway that sits between internal product teams and multiple model providers/backends, handling routing, streaming, quotas, and safety...

    OpenAI / Anthropic
    Free
  6. 06

    AI Training Cluster

    Design the scheduler for a large GPU/TPU training cluster shared by multiple teams running distributed training jobs, handling gang scheduling, preemption, and...

    Meta
    Preview
  7. 07

    MCP Tool Gateway

    Design an MCP-style tool gateway that lets AI agents discover and invoke enterprise tools (CRM writes, ticketing, internal APIs) safely, with per-user authoriza...

    Anthropic
    Preview
  8. 08

    AI Customer Support Agent

    Design an AI customer support agent for a SaaS product.

    Practice bank
    Free
  9. 09

    Ad Ranking Evaluation Framework

    Design an evaluation framework for an ad ranking system.

    Meta
    Preview
  10. 10

    ML Feature Store

    Design an ML feature store that serves the same features to both offline model training and low-latency online inference for a company like Uber.

    Uber
    Preview
  11. 11

    Vector Database

    Design the storage and query engine of a vector database that indexes billions of embeddings for approximate nearest-neighbor search.

    Meta (Faiss)
    Preview
  12. 12

    ML Model Serving Platform

    Design a general-purpose ML model serving platform for non-LLM prediction serving used by many teams.

    NVIDIA
    Preview
  13. 13

    News Feed Ranking

    Design the ranking system for a social news feed like Instagram or Facebook.

    Meta (Instagram)
    Preview
  14. 14

    Fraud Detection

    Design a real-time ML fraud/risk scoring system that scores payment transactions in under 100 ms.

    Stripe
    Preview
  15. 15

    ETA Prediction

    Design an ETA prediction service that returns arrival-time estimates for millions of trips per second.

    Uber
    Preview
  16. 16

    LLM Guardrails Pipeline

    Design a guardrails pipeline that wraps an LLM application with layered input and output safety.

    NVIDIA / Meta
    Preview
  17. 17

    LLM Evaluation Platform

    Design the evaluation platform for an organization shipping LLM features: manage a versioned golden set (stratified production samples, an adversarial library,...

    Practice bank
    Preview
  18. 18

    Real-time Voice AI Agent

    Design a real-time voice AI agent platform for phone support and voice assistants: the complete loop from the user speaking to the agent speaking back.

    Practice bank
    Preview