Back to HLD Sheets & Companies
Google System Design Track

Google High-Level Design (HLD) Interview Questions

Global-scale storage, consensus, search indexing & real-time collaboration. Practice each prompt in HLDArena's 60-second discrete-event simulator with p99 latency, availability, and cloud cost gates.

Track Progress0 / 22 Solved (0%)
What Google Loops Probe Deepest
  • Multi-region consistency, Spanner/Paxos quorum trade-offs & RPO = 0 durability
  • Inverted indexes, typeahead tries, and crawl frontier partitioning
  • Real-time OT/CRDT operation sequencing & chunked document storage
StatusCodeQuestionDifficultyMax Pts
—HLD-407Global Multiplayer Document Revision Log & OT/CRDT Sequencer

Design Google Docs / Notion’s collaborative editing backend sequencing concurrent character-level OT/CRDT operations, live comments, and lossless revision history.

Expert2000 pts
—HLD-409Planet-Scale Web Search Inverted Index & Query Aggregator

Design Google / Bing’s real-time web search serving tier combining tiered result caches, term/document-partitioned inverted index leaves, and breaking-news fresh index ingestion.

Expert2000 pts
—HLD-209Distributed Web Crawler URL Frontier & Dedup

Design a URL discovery, deduplication, and politeness-throttled crawl frontier coordinating thousands of concurrent fetcher workers.

Medium1000 pts
—HLD-202Product Catalog & Autocomplete Typeahead Search

Design an e-commerce prefix autocomplete and SKU lookup backend serving keystroke-by-keystroke suggestions within strict sub-40ms latency budgets during holiday sales.

Medium1000 pts
—HLD-235Cloud Photo Backup, Dedup & Timeline Album Service

Design Google Photos / iCloud Photos metadata timeline, SHA-256 content deduplication, and shared event album delivery across mobile devices.

Medium1000 pts
—HLD-226Enterprise Calendar Scheduling, Recurrence & Sync

Design Google Calendar’s core event storage, RRULE recurrence expansion, and attendee calendar synchronization handling top-of-hour reminder bursts.

Medium1000 pts
—HLD-405Multi-Tenant Distributed Lock & Lease Coordinator

Design a strongly consistent distributed lock and monotonic fencing-token service backing leader election across thousands of tenant clusters.

Expert2000 pts
—HLD-411Geo-Partitioned Distributed SQL & Serializable Transaction Engine

Design a Spanner / CockroachDB-inspired range-partitioned transactional database proxy supporting multi-key serializable commits, range splits, and leader lease failover.

Expert2000 pts
—HLD-408Multi-Cluster LLM Prefix KV-Cache Router & Token Scheduler

Design an OpenAI / vLLM multi-cluster inference control plane that routes shared system-prompt prefixes to sticky KV-cache pools, enforces TPM quotas, and survives 8x storage/checkpoint brownouts.

Expert2000 pts
—HLD-128Real-Time Phishing & Safe-Browsing URL Reputation API

Design a Google Safe Browsing-style URL prefix reputation service queried by browsers and email gateways before rendering external links.

Easy500 pts
—HLD-229Distributed Tracing Span Collector & Tail Sampler

Design a Dapper / Jaeger / Datadog APM tracing backend that reassembles microservice spans by trace_id for tail-based error sampling and waterfall visualization.

Medium1000 pts
—HLD-126Calendar Free/Busy Slot Lookup & Booking Link API

Design a Calendly-style public scheduling availability API that serves precomputed free/busy bitmaps when hosts share booking links widely.

Easy500 pts
—HLD-217Sticky A/B Experiment Assignment & Exposure Log

Design an experimentation bucketing and sticky-override service ensuring users always see a consistent treatment arm across requests while logging exposures.

Medium1000 pts
—HLD-231Enterprise RAG Document Chunking & Hybrid Vector Search

Design a Glean / Enterprise AI retrieval backend combining ACL-filtered BM25 keyword search + dense vector embeddings with semantic query caching.

Medium1000 pts
—HLD-213Video Upload & Multi-Bitrate HLS Transcode Queue

Design a video upload metadata and transcode orchestration control plane dispatching chunked encoding tasks and serving studio readiness status.

Medium1000 pts
—HLD-107Distributed 64-Bit ID & Sequence Service

Design a high-throughput unique ID allocation service issuing time-ordered 64-bit identifiers and range leases to internal microservices.

Easy500 pts
—HLD-224Real-Time Top-K Trending Topics & Bestsellers Engine

Design an Amazon Bestsellers / X Trending heavy-hitters pipeline computing sliding-window Top-K rankings (15m / 1h / 24h) from high-velocity interaction streams.

Medium1000 pts
—HLD-121Microservice Health-Check & Service Discovery Registry

Design a Consul / Netflix Eureka-style service registry handling continuous pod heartbeats and high-QPS endpoint discovery lookups across thousands of microservices.

Easy500 pts
—HLD-321Global Video Conferencing Room Signaling & SFU Topology

Design Zoom / Google Meet’s room signaling, participant roster state, and regional Selective Forwarding Unit (SFU) relay allocation at 9:00 AM join storms.

Hard1500 pts
—HLD-322Sharded HNSW Vector Database & Metadata Filter Engine

Design a Pinecone / Milvus multi-tenant vector database serving low-latency ANN top-K cosine similarity queries with real-time vector upserts and deletes.

Hard1500 pts
—HLD-225Centralized Log Ingestion & Full-Text Search Pipeline

Design a Datadog Logs / ELK-style observability pipeline ingesting structured JSON logs at high write throughput while serving incident search queries.

Medium1000 pts
—HLD-239Global Airline Fare Cache & Multi-Leg Search Aggregator

Design a Google Flights / Skyscanner search backend serving sub-60ms route-date fare matrices while airline GDS feeds continuously push seat-class repricing.

Medium1000 pts
Open Must-Do 30 Sheet