Google

Solve a problem (pass its tests) and it's checked off here automatically — your progress is saved, so it stays checked when you come back.

• BatchNorm forward (train + eval modes)NormalizationRegularizationMedium
• BERTScore (Greedy Token-Cosine F1)NaturalLanguageProcessingEvaluationMetricsEmbeddingsMedium
• Best-of-N SelectionLargeLanguageModelEvaluationMetricsEasy
• Binary Quantization + Hamming RetrievalNeuralNetworksLargeLanguageModelEmbeddingsMedium
• BM25 ScoringNaturalLanguageProcessingLargeLanguageModelMLSystemDesignMedium
• Bootstrap Confidence Interval (Percentile)StatisticsMedium
• Bradley-Terry Strength MLEProbabilityStatisticsMLSystemDesignHard
• Brier Score + Reliability DecompositionEvaluationMetricsMedium
• Scaled dot-product self-attentionTransformersLargeLanguageModelMedium
• Linear forward (Wx + b)NeuralNetworksEasy
• Chi-Square Test of IndependenceStatisticsMedium
• Clipped Inverse Propensity WeightingCalculusMLSystemDesignMedium
• Cost-Sensitive Decision ThresholdProbabilityEvaluationMetricsMedium
• Critic Verdict ParserAgenticAIEasy
• Cross-Attention (Encoder-Decoder)TransformersMedium
• DDM Concept Drift DetectorStatisticsMLSystemDesignMedium
• DDPM Reverse (Denoise) StepNeuralNetworksMedium
• Delta Method for Ratio Metric VarianceStatisticsHard
• Difference-in-DifferencesStatisticsMedium
• Elo Rating UpdateProbabilityStatisticsMLSystemDesignEasy
• Expected Calibration Error (ECE)EvaluationMetricsMedium
• Expected Reciprocal Rank (ERR)MLSystemDesignEvaluationMetricsMedium
• Feature Hashing (Hashing Trick)MLSystemDesignMedium
• Fleiss' KappaStatisticsEvaluationMetricsMedium
• Group / Time-Series CV SplitComputerVisionMLSystemDesignEasy
• Hard Negative Mining for RetrieverNeuralNetworksMLSystemDesignEmbeddingsMedium
• Hash Bucket A/B AssignmentStatisticsEasy
• HNSW Greedy Graph SearchNeuralNetworksMLSystemDesignHard
• Post-Training int8 QuantizationLargeLanguageModelMLSystemDesignMedium
• Team-Draft Interleaving (Online Ranking Eval)StatisticsMLSystemDesignMedium
• Isolation Forest Anomaly ScoreStatisticsDecisionTreesMedium
• Isotonic Regression (PAVA)EvaluationMetricsHard
• IVF (Inverted File) SearchNeuralNetworksMLSystemDesignMedium
• Jensen-Shannon Divergence (Drift)ProbabilityMLSystemDesignMedium
• JSON Tool-Arg RepairAgenticAIMedium
• Kendall's TauStatisticsMLSystemDesignMedium
• Knowledge-Distillation LossMLSystemDesignLossFunctionsFineTuningMedium
• Kolmogorov-Smirnov StatisticStatisticsMLSystemDesignMedium
• Label-Noise Detection (Confident Learning)NeuralNetworksMLSystemDesignHard
• LambdaMART Lambda GradientsMLSystemDesignEvaluationMetricsHard
• Levenshtein Edit DistanceNaturalLanguageProcessingEasy
• LinUCB Contextual BanditStatisticsAgenticAIHard
• ListNet Listwise Ranking LossMLSystemDesignLossFunctionsMedium
• Little's LawMLSystemDesignOverfittingEasy
• Longest Common SubsequenceNaturalLanguageProcessingMedium
• Magnitude PruningMLSystemDesignEasy
• Mann-Whitney U TestStatisticsMLSystemDesignMedium
• mAP for Object DetectionComputerVisionEvaluationMetricsHard
• Sentence Embedding (Masked Mean-Pool)NaturalLanguageProcessingMLSystemDesignEmbeddingsEasy
• Matryoshka Embedding TruncationMLSystemDesignEmbeddingsMedium
• Matthews Correlation Coefficient (MCC)EvaluationMetricsEasy
• Mean Average Precision (MAP)MLSystemDesignEvaluationMetricsMedium
• MinHash for Near-Duplicate DetectionMLSystemDesignMedium
• MIPS -> L2 ReductionNeuralNetworksLinearAlgebraMLSystemDesignMedium
• Maximal Marginal Relevance (MMR)LargeLanguageModelMLSystemDesignMedium
• Multi-Task Score FusionMLSystemDesignTransferLearningMedium
• Multi-Tool Router (by Embedding)MLSystemDesignAgenticAIMedium
• Multiple-Testing Correction (Bonferroni / BH-FDR)StatisticsMedium
• Negative Sampling (Implicit Feedback)StatisticsMLSystemDesignMedium
• Out-of-Fold Target EncodingMLSystemDesignMedium
• Pairwise Ranking Loss (BPR / RankNet)MLSystemDesignLossFunctionsMedium
• pass@k Unbiased EstimatorStatisticsLargeLanguageModelEvaluationMetricsMedium
• Permutation Feature ImportanceMLSystemDesignEvaluationMetricsMedium
• PII Detection & RedactionNaturalLanguageProcessingMLSystemDesignAgenticAIMedium
• Pinball (Quantile) Regression LossLinearRegressionStatisticsLossFunctionsMedium
• Plan-DAG Cycle DetectionAgenticAIMedium
• Planner Output ParserNaturalLanguageProcessingAgenticAIEasy
• Platt ScalingLogisticRegressionEvaluationMetricsMedium
• Point-in-Time Feature JoinStatisticsMLSystemDesignMedium
• Population Stability Index (PSI)StatisticsMLSystemDesignMedium
• PR-AUC (Average Precision)EvaluationMetricsMedium
• Prefix-Cache Hit AccountingLargeLanguageModelMLSystemDesignAgenticAIEasy
• Product Quantization (PQ) Encode + ADCNeuralNetworksLargeLanguageModelMLSystemDesignHard
• Prompt-Injection HeuristicAgenticAIMedium
• QA Token F1 / Exact MatchNaturalLanguageProcessingEvaluationMetricsMedium
• Random-Hyperplane LSH (Cosine)NeuralNetworksMLSystemDesignEmbeddingsMedium
• ReAct Loop DriverAgenticAIMedium
• Recall@k (Retrieval)LargeLanguageModelMLSystemDesignEvaluationMetricsEasy
• Reciprocal Rank Fusion (RRF)LargeLanguageModelMLSystemDesignMedium
• Reflexion Retry With FeedbackAgenticAIMedium
• Risk-Coverage Curve (AURC)EvaluationMetricsMedium
• Run-Level Cost / Budget CeilingMLSystemDesignAgenticAIMedium
• Sample Ratio Mismatch (SRM) CheckStatisticsMLSystemDesignMedium
• Sample Size for Statistical PowerStatisticsMedium
• In-Batch Sampled Softmax + logQ CorrectionNeuralNetworksMLSystemDesignLossFunctionsHard
• Self-Consistency Majority VoteLargeLanguageModelEvaluationMetricsEasy
• Semaphore-Bounded Concurrent TasksMLSystemDesignAgenticAIMedium
• Sequential Test (mSPRT-Style Always-Valid p-Value)StatisticsHard
• Shadow / Canary Rollback DecisionMLSystemDesignAgenticAIMedium
• Shared-Memory Tool-Call CacheAgenticAIMedium
• Sliding-Window Counter (Streaming Feature)MLSystemDesignMedium
• Smoothed Target (Mean) EncodingMLSystemDesignMedium
• SMOTE OversamplingStatisticsEvaluationMetricsMedium
• Softmax from scratchNeuralNetworksEasy
• Spearman Rank CorrelationStatisticsMLSystemDesignEvaluationMetricsEasy
• Task-DAG Parallel SchedulerMLSystemDesignAgenticAIHard
• Temporal Train/Val SplitStatisticsMLSystemDesignEasy
• Thompson Sampling (Bernoulli-Beta)ProbabilityStatisticsAgenticAIMedium
• Token-Bucket Rate LimiterMLSystemDesignEasy
• Token-Budget Context ManagerLargeLanguageModelAgenticAIMedium
• Tool-Call Parse + Schema ValidationEvaluationMetricsAgenticAIMedium
• Trace Aggregation SummaryMLSystemDesignAgenticAIEasy
• Two-Sample Test + p-valueStatisticsMedium
• Two-Tower Top-k RetrievalMLSystemDesignEmbeddingsEasy
• ViT Patch EmbeddingTransformersComputerVisionEmbeddingsMedium
• Wilson Confidence Interval (Proportion)StatisticsEasy
• WordPiece Tokenization (Greedy Longest-Match)NaturalLanguageProcessingMLSystemDesignMedium