Questions
Status
Point-in-time feature correctnessHardDesignStripe styleDesignStripe styleHard50 minMulti-model serving platformHardDesignOpenAI styleDesignOpenAI styleHard45 minDrift detection and auto-retrainHardDesignWalmart styleDesignWalmart styleHard45 minEmbedding version migrationHardDesignMeta styleDesignMeta styleHard50 minReproducible regulated model trainingHardDesignCapital One styleDesignCapital One styleHard45 minStreaming windowed aggregationsHardDesignPayPal styleDesignPayPal styleHard50 minGPU-shared model cold startHardDesignOpenAI styleDesignOpenAI styleHard45 minOnline feed ranking updatesHardDesignMeta styleDesignMeta styleHard50 minMonitoring with delayed ground truthHardDesignCapital One styleDesignCapital One styleHard45 minETA prediction serviceHardDesignGoogle styleDesignGoogle styleHard50 minVideo scene search and indexingHardDesignGoogle styleDesignGoogle styleHard50 minLive competitive game leaderboardHardDesignDesignHard45 minSocial feed rankingHardDesignMeta styleDesignMeta styleHard55 minFeature store online and offlineHardDesignUber styleDesignUber styleHard55 minMulti-tenant model servingHardDesignNvidia styleDesignNvidia styleHard55 minDaily CTR model training pipelineHardDesignGoogle styleDesignGoogle styleHard55 minML experimentation platformHardDesignNetflix styleDesignNetflix styleHard50 minReal-time fraud detectionHardDesignStripe styleDesignStripe styleHard55 minAds bidding and rankingHardDesignGoogle styleDesignGoogle styleHard55 minContent moderation pipelineHardDesignMeta styleDesignMeta styleHard55 minMarketplace search rankingHardDesignAmazon styleDesignAmazon styleHard55 minModel drift detectionHardDesignDatabricks styleDesignDatabricks styleHard50 minTwo-stage video recommenderHardDesignMeta styleDesignMeta styleHard55 minFraud scoring with adaptive featuresHardDesignStripe styleDesignStripe styleHard50 minPayment fraud scoring systemHardDesignStripe styleDesignStripe styleHard45 minMulti-armed bandit experimentation platformHardDesignNetflix styleDesignNetflix styleHard40 minFeature and prediction drift detectionHardDesignDatabricks styleDesignDatabricks styleHard40 minModel deployment and canary rolloutHardDesignDatabricks styleDesignDatabricks styleHard45 minDiscovery feed rankingHardDesignMeta styleDesignMeta styleHard55 minReal-time fraud feature storeHardDesignStripe styleDesignStripe styleHard45 minAd CTR prediction and calibrationHardDesignGoogle styleDesignGoogle styleHard45 minGraph-based fraud featuresHardDesignStripe styleDesignStripe styleHard45 min
Ranking experimentation platformHardDesignGoogle styleDesignGoogle styleHard50 minRAG backend with access controlHardDesignOpenAI styleDesignOpenAI styleHard50 minDistributed LLM trainingHardDesignOpenAI styleDesignOpenAI styleHard50 minLLM inference servingHardDesignOpenAI styleDesignOpenAI styleHard45 minRanking model rollout safetyHardDesignGoogle styleDesignGoogle styleHard40 minCoding agent retrieval cachingHardDesignAnthropic styleDesignAnthropic styleHard50 minGNN serving at scaleHardDesignMeta styleDesignMeta styleHard45 minABAC policy decision engineHardDesignAmazon styleDesignAmazon styleHard45 minBatch photo processing pipelineHardDesignGoogle styleDesignGoogle styleHard45 minInverted-index posting-list evaluationHardDesignGoogle styleDesignGoogle styleHard45 minDesign offline ranker evaluationHardDesignMeta styleDesignMeta styleHard50 minFresh document RAG retrievalHardDesignOpenAI styleDesignOpenAI styleHard50 minContent safety inference at scaleHardDesignMeta styleDesignMeta styleHard45 minInterleaved ranking experimentsHardDesignNetflix styleDesignNetflix styleHard45 minLLM prefix KV cache sharingHardDesignOpenAI styleDesignOpenAI styleHard50 minLLM agent quality monitoringHardDesignAnthropic styleDesignAnthropic styleHard45 minComplementary item retrievalHardDesignAmazon styleDesignAmazon styleHard45 minDistributed training fault toleranceHardDesignOpenAI styleDesignOpenAI styleHard50 min