LinkedIn ML & AI interview questions.
Questions in the CrackML corpus associated with LinkedIn. Use company evidence as a prioritization signal, not a promise that an exact question will repeat.
Explain the attention mechanism in a Transformer.GenAI & LLM · medium · Evidence 78/100Partition into K equal-sum subsetsPython & DSA · hard · Evidence 77/100Sample From a Non-Uniform RNGML Coding & PyTorch · hard · Evidence 77/100Implement rejection samplingML Coding & PyTorch · medium · Evidence 77/100Design a cold-start recommenderML System Design · hard · Evidence 77/100How would you handle an imbalanced dataset?ML Fundamentals · medium · Evidence 76/100Dropout vs weight decayML Fundamentals · medium · Evidence 47/100Project Deep Dive: Define the ProblemProject & Behavioral · medium · Evidence 41/100Explain embeddings, tokenization, attention, and BERTGenAI & LLM · medium · Evidence 41/100How do tokenization choices affect professional-domain text?GenAI & LLM · medium · Evidence 41/100BERT: Architecture and TrainingML Fundamentals · medium · Evidence 41/100Model Explainability in ProductionML Fundamentals · medium · Evidence 41/100When would you combine topic modeling with embedding-based clustering?ML Fundamentals · medium · Evidence 41/100When would BERT be overkill compared with a simpler embedding model?GenAI & LLM · medium · Evidence 41/100Implement Cross-Entropy Loss From ScratchML Coding & PyTorch · medium · Evidence 41/100Design a LinkedIn Classification SystemML System Design · hard · Evidence 41/100Explain Technical Work to a Non-Technical StakeholderProject & Behavioral · medium · Evidence 41/100Choosing a Clustering AlgorithmML Fundamentals · medium · Evidence 41/100Ambiguous RequirementsProject & Behavioral · hard · Evidence 41/100Derive Binary Cross-EntropyML Math · medium · Evidence 41/100Implement a Contrastive Learning LossML Coding & PyTorch · hard · Evidence 41/100Project Deep Dive: Outcome and TakeawayProject & Behavioral · medium · Evidence 41/100Contrastive Learning for RecommendationGenAI & LLM · hard · Evidence 41/100How would you generate candidates for a brand-new user before personalization data exists?ML System Design · medium · Evidence 41/100Design a Recommendation System at High LevelML System Design · hard · Evidence 41/100Transformation + Sorting AI Coding ProblemML Coding & PyTorch · medium · Evidence 41/100When would you choose topic modeling instead of generic clustering?ML Fundamentals · medium · Evidence 41/100Embeddings for Retrieval vs Token EmbeddingsGenAI & LLM · medium · Evidence 41/100How would you avoid feedback loops in a professional recommendation system?ML System Design · hard · Evidence 41/100How should retrieval metrics differ from ranking metrics in a recommender?ML Fundamentals · medium · Evidence 41/100Design Recommendation + Ranking for a Hiring-Manager RoundML System Design · hard · Evidence 41/100Defend an ML project end to endProject & Behavioral · hard · Evidence 41/100What negative-sampling strategy would you use for implicit-feedback recommendation?ML Fundamentals · hard · Evidence 41/100Bayes Theorem for a Rain Prediction ScenarioML Math · medium · Evidence 41/100Tokenization TradeoffsML Fundamentals · medium · Evidence 41/100How would you handle sparse implicit feedback in a cold-start recommender?ML System Design · hard · Evidence 41/100How would you engineer features from sparse professional profiles?ML System Design · medium · Evidence 41/100SVM Margin and Hinge LossML Math · medium · Evidence 41/100Cold Start: New Users vs New ItemsML System Design · medium · Evidence 41/100How do dropout and weight decay interact as regularizers?ML Fundamentals · hard · Evidence 41/100How would you evaluate clustering quality for unlabeled user segments?ML Fundamentals · medium · Evidence 41/100How would you prove that a rejection-sampling procedure is correct?ML Math · hard · Evidence 41/100Project Deep Dive: How Did You Design the Solution?Project & Behavioral · hard · Evidence 41/100What Makes a Good Embedding?ML Fundamentals · medium · Evidence 41/100Retrieval vs RankingML Fundamentals · medium · Evidence 41/100How would you separate offline and online pipelines in a large recommender?ML System Design · hard · Evidence 41/100Topic Modeling: LDA vs Embedding ClustersML Fundamentals · medium · Evidence 41/100How does attention complexity affect long-document modeling?GenAI & LLM · hard · Evidence 41/100