Design a Logging and Data Ingestion System
Accept high-volume log streams, durably buffer them, make recent and historical data queryable, and tail newly arriving records to authorized readers.
Open practiceMicrosoft interview practice
Practice from 247 Microsoft-tagged public coding problems, organized only by the stage, topic, difficulty, and recency metadata FastPrep can verify. The broader public catalog also includes 35 system-design exercises, 5 low-level-design exercises.
OA practice path
Build a timed implementation rhythm from prompt to edge cases. Problems are ranked by repeated public catalog sightings.
111 Microsoft-tagged Online Assessment problems available.
| Company | Problem | Difficulty | Public evidence | Action |
|---|---|---|---|---|
MIMicrosoft | Maximum Escape Game ScoreArrayHash Table | Medium | 6 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Minimum Non-Adjacent Selection CapabilityArrayBinary Search | Medium | 5 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Authentication SystemHash TableDesign | Medium | 5 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Maximum Valid Substring FrequencyStringHash Table | Medium | 4 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Maximum Reward PointsArrayGreedy | Medium | 4 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Minimum Cost K-Capable ModelsArrayGreedy | Medium | 3 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | K Most Recent Unique Request IDsArrayHash Table | Easy | 3 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Minimum Fuel Cost Between CitiesGraphShortest Path | Hard | 3 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | XOR MultiplicationBit ManipulationMath | Medium | 3 public reportsLast reported Jul 2026 | Practice |
MIMicrosoft | Maximum Interactive Team SizeArraySorting | Medium | 2 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Maximize Pipeline Throughput Under a Scaling BudgetBinary SearchGreedy | Medium | 2 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Prepare NotificationStringGreedy | Easy | 2 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Shortest Board Length to Cover HolesSortingBinary Search | Medium | 2 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Shopkeeper Final Price SummaryArrayStack | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Minimum Jump CostArrayDynamic Programming | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Maximum Decrypted ScoreArraySliding Window | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Minimum Team Size From Every StartArraySliding Window | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Maximum String OperationsStringGreedy | Medium | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | 0/1 KnapsackDynamic ProgrammingArray | Medium | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Maximum Score with Prime JumpsArrayDynamic Programming | Medium | 2 public reportsLast reported Aug 2026 | Practice |
Phone practice path
Practice solving while explaining assumptions, tradeoffs, and complexity. Problems are ranked by repeated public catalog sightings.
| Company | Problem | Difficulty | Public evidence | Action |
|---|---|---|---|---|
MIMicrosoft | Least Recently Used CacheHash TableDesign | Medium | 4 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Retain Top K ValuesArraySorting | Easy | 4 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Deterministic K-Means AssignmentsArrayMath | Hard | 4 public reportsLast reported May 2026 | Practice |
MIMicrosoft | Greedy and Beam Sentence DecodingHeapGreedy | Hard | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | 3SumArraySorting | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Arithmetic Expression EvaluatorStringStack | Hard | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Infer a Lexicographically Smallest Alien AlphabetStringGraph | Hard | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Sorted Subsets of a StringStringBacktracking | Medium | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Top K Frequent ElementsHash TableHeap | Medium | 1 public reportLast reported Sep 2026 | Practice |
MIMicrosoft | Employee Hierarchy Cycle and Depth AnalysisGraphDepth-First Search | Medium | 1 public reportLast reported Sep 2026 | Practice |
MIMicrosoft | Most Visited String SectorsArraySimulation | Easy | 1 public reportLast reported Sep 2026 | Practice |
MIMicrosoft | Suffix Maximum Frequency QueriesArrayDynamic Programming | Easy | 1 public reportLast reported Aug 2026 | Practice |
MIMicrosoft | Lowest Common Ancestor in Ordered and Unordered TreesTreeDepth-First Search | Medium | 1 public reportLast reported Aug 2026 | Practice |
MIMicrosoft | Rank Open BusinessesArraySorting | Easy | 1 public reportLast reported May 2026 | Practice |
MIMicrosoft | Min StackStackDesign | Medium | 1 public reportLast reported Apr 2026 | Practice |
MIMicrosoft | Excel Training Data and Model Response AuditArrayMatrix | Easy | 1 public reportLast reported Mar 2026 | Practice |
MIMicrosoft | Word Break II: All SentencesStringDynamic Programming | Hard | 1 public reportLast reported Jan 2026 | Practice |
MIMicrosoft | Three Partition ArrayArrayTwo Pointers | Easy | 1 public reportLast reported Dec 2024 | Practice |
MIMicrosoft | Sort Tuples by Value and Then KeySortingArray | Easy | 1 public reportLast reported Apr 2024 | Practice |
MIMicrosoft | Target-Sum Pair with Two PointersArraySorting | Easy | 1 public reportLast reported Apr 2024 | Practice |
Onsite practice path
Practice complete solutions and defend the choices behind them. Problems are ranked by repeated public catalog sightings.
| Company | Problem | Difficulty | Public evidence | Action |
|---|---|---|---|---|
MIMicrosoft | Least Recently Used CacheHash TableDesign | Medium | 4 public reportsLast reported Oct 2026 | Practice |
MIMicrosoft | Retain Top K ValuesArraySorting | Easy | 4 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Deterministic K-Means AssignmentsArrayMath | Hard | 4 public reportsLast reported May 2026 | Practice |
MIMicrosoft | Build a Precision-Recall CurveArraySorting | Medium | 3 public reportsLast reported Nov 2025 | Practice |
MIMicrosoft | Greedy and Beam Sentence DecodingHeapGreedy | Hard | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | 3SumArraySorting | Medium | 2 public reportsLast reported Sep 2026 | Practice |
MIMicrosoft | Infer a Lexicographically Smallest Alien AlphabetStringGraph | Hard | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | 0/1 KnapsackDynamic ProgrammingArray | Medium | 2 public reportsLast reported Aug 2026 | Practice |
MIMicrosoft | Merge IntervalsIntervalsSorting | Medium | 2 public reportsLast reported Apr 2026 | Practice |
MIMicrosoft | Is Graph Bipartite?GraphBreadth-First Search | Medium | 2 public reportsLast reported Feb 2022 | Practice |
MIMicrosoft | Number of IslandsMatrixGraph | Medium | 2 public reportsLast reported Oct 2021 | Practice |
MIMicrosoft | Compute Population VarianceArrayMath | Medium | 2 public reportsLast reported Mar 2021 | Practice |
MIMicrosoft | Generate Fixed-Size CombinationsBacktrackingCombinatorics | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Match a Descendant Selector in a DOM TreeTreeDepth-First Search | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Capacity to Ship Packages Within DaysArrayBinary Search | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Copy a Doubly Linked List with a Special PointerLinked ListHash Table | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Amount of Time to Infect a TreeGraphBreadth-First Search | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | K-Second Message SuppressionArrayHash Table | Easy | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Max Consecutive Ones IIIArraySliding Window | Medium | 1 public reportLast reported Oct 2026 | Practice |
MIMicrosoft | Maximum Subarray Sum Across K ConcatenationsArrayDynamic Programming | Medium | 1 public reportLast reported Oct 2026 | Practice |
01 · Preparation plan
The Microsoft catalog is broad enough to reward deliberate rotation. Start with a parser, move to a dependency problem, and then reason about state over time. Use the separate design collection to extend those implementation choices into service boundaries; do not treat the largest stage group as a prediction of your interview. This is a suggested practice sequence, not the employer's interview process. Your invitation and recruiter guidance remain the source of truth.
Use the expression evaluator to separate token recognition from operator precedence. A clear grammar makes nested input easier to reason about than a collection of special cases.
The employee hierarchy exercise is useful for distinguishing a valid tree from malformed reporting relationships. Explain why cycle detection must happen before computing a reliable depth.
Treat the authentication-system exercise as a contract for issuing, renewing, and expiring state. After implementation, use the public service-design exercises to discuss where that state belongs.
02 · Broader technical practice
These public Microsoft-tagged exercises cover additional technical formats. They are included only when a verified catalog record and a crawlable practice page both exist.
Accept high-volume log streams, durably buffer them, make recent and historical data queryable, and tail newly arriving records to authorized readers.
Open practiceDesign event ingestion, candidate retrieval, ranking, and serving for a bounded personalized user-item catalog.
Open practiceMatch two players and execute authoritative turn-based games through retry-safe APIs across multiple regions.
Open practiceDesign a single-retailer shopping API whose login, session, cart, and order boundaries remain secure and retry-safe at scale.
Open practiceDesign company-managed product pages with trackable share links, anonymous milestone rewards, and trustworthy engagement analytics.
Open practiceModel infrastructure assets and typed relationships, then validate and construct traversable attack paths.
Open practiceIngest HR snapshots, train an imbalance-aware attrition model, and deliver auditable explanations to authorized reviewers.
Open practiceMonitor a geographically distributed lab-machine fleet with timely health state, alerts, history, and partition-aware recovery.
Open practiceDesign a permission-aware RAG service that answers questions over a large document vault with verifiable citations.
Open practiceDesign a reliable multi-channel notification platform for a high-cardinality internal producer fleet, with durable event handoff, fair admission, provider isolation, and recoverable poison messages.
Open practiceCoordinate local desktop notifications from multiple applications with durable intent, user policy, bounded delivery, and restart-safe state.
Open practiceDesign a location search service that returns nearby restaurants in stable distance order.
Open practiceDesign a Drive-like service for durable, chunked file storage, sharing, versions, and multi-device synchronization. Audit a first-letter photo placement plan and its database-cost drivers.
Open practiceDesign a voice receptionist that conducts bounded conversations and hands calls to people or workflows.
Open practiceDesign a globally available service for reliable one-to-one internet voice calls.
Open practiceShard very large input-embedding and language-model output matrices while controlling memory, communication, skew, and sampling cost.
Open practiceModel idempotent vehicle entry and exit events, one open journey per vehicle, deterministic toll calculation, and a path to partitioned durable processing.
Open practiceDesign a high-concurrency service that connects a support ticket to related user tickets and the affected system's historical status.
Open practiceDesign a large-marketplace search service that filters by price and taxonomy, serves stable ranked pages, and safely incorporates versioned catalog changes.
Open practiceIssue, deliver, and verify short-lived one-time passwords while isolating tenant policy, abuse, and data.
Open practiceContinuously crawl an allowed web corpus and build a fresh, deduplicated inverted index for keyword search.
Open practiceDesign a multi-tenant backend that ingests versioned documents and serves low-latency text search with filters, stable pagination, and safe index rebuilds.
Open practiceDesign a partitioned cache with concurrent access, per-entry expiration, and approximate local recency eviction.
Open practiceRecommend timely local sports coverage with retrieval-grounded summaries, safe fallbacks, and measurable model cost and quality.
Open practiceDesign a desktop notification center with ordering, dismissal, and per-application state.
Open practiceDesign and implement a library circulation system with titles, physical copies, members, loans, holds, extensible policies, and atomic checkout and return behavior.
Open practiceDesign a multi-region multiplayer game that matches players and runs low-latency authoritative sessions.
Open practiceImprove an assistant's factual and abstention consistency across equivalent prompts with reviewed SFT data, reward design, and independent evaluation.
Open practiceDesign the data, temporal representation, joint training and independent evaluation needed to build a first video-language question-answering model.
Open practiceInfer a safe, calibrated task intent from changing partial input without confusing intent with text completion.
Open practiceDesign an online ranker and its model lifecycle for caller-supplied candidates.
Open practiceTurn an analyst's natural-language question into a grounded, authorized Kusto query with a safe preview path.
Open practiceImprove the latency and unit cost of an existing streaming LLM service while protecting answer quality and safe rollout.
Open practiceAutomatically compare reproducible tabular classification pipelines, select a model, and enforce a finite search budget despite retries and worker failures.
Open practiceClassify websites from domain and page signals while measuring crowd-label quality and protecting minority classes.
Open practiceLearn query-to-link relevance from historical clicks while separating weak feedback, candidate coverage and safe online ranking.
Open practiceRetrieve and rank relevant text and image ads for a query while learning safely from sparse, biased feedback.
Open practiceBuild a bounded in-memory cache that preserves exact least-recently-used eviction when several threads share it.
Open practiceBuild, evaluate, and release an object-detection model for cars, bicycles, and trucks under severe scenario and class imbalance.
Open practicePredict useful next spreadsheet commands from an ordered session history without leaking future actions or exposing private workbook content.
Open practice03 · Evidence boundary
It means practicing transferable implementation, testing, and technical reasoning with public catalog assets FastPrep tags to Microsoft. It does not mean FastPrep has access to the company's assessments or any private interview bank.
Repeated public sightings and last-reported dates can help you prioritize practice, but they cannot predict the questions, format, or platform in a specific interview.
04 · Plain answers
This page owns technical-practice intent for Microsoft. Hiring activity, timelines, and market signals remain on the separate hiring-insights page.
No. The available practice spans timed coding, explanation-heavy problems, and design exercises. Use assessment-labeled records for implementation rehearsal, then switch to dependencies and stateful APIs when your role calls for discussion. This is a software-engineering study path, not a guide to every Microsoft role or a fixed sequence of rounds.
No. FastPrep is an independent interview-preparation product and is not affiliated with Microsoft. The page uses FastPrep's public practice catalog and does not claim official, private, leaked, or proprietary employer questions.
Stage labels come from FastPrep's public problem metadata. A problem can carry more than one reported stage, and hiring processes can change by role, level, location, and date. Treat the labels as preparation context, not a guarantee.
Start with the stage named in your invitation or recruiter message. If no stage is known, use the largest available set to build general problem-solving fluency, then rehearse explanation and testing separately.
No. Public catalog counts, stage tags, and last-reported dates can help prioritize practice, but they cannot predict a specific interview's questions, sequence, timing, or platform.
Your next stage, made concrete
Start with the stage named in your invitation, then use public evidence as context—not as a promise of what you will be asked.
Open the practice set