TikTok ML & AI interview questions.
Questions in the CrackML corpus associated with TikTok. Use company evidence as a prioritization signal, not a promise that an exact question will repeat.
Explain Transformer architecture and each major component.GenAI & LLM · medium · Evidence 88/100Serialize and Deserialize Binary TreePython & DSA · hard · Evidence 80/100Explain the attention mechanism in a Transformer.GenAI & LLM · medium · Evidence 78/100Explain the bias-variance trade-off.ML Fundamentals · easy · Evidence 77/100How would you handle an imbalanced dataset?ML Fundamentals · medium · Evidence 76/100How would you improve detection of very small objects?ML Fundamentals · hard · Evidence 75/100How do learning-rate schedules affect training?ML Fundamentals · medium · Evidence 75/100Find and print a path from a source to a destination through obstacles.Python & DSA · medium · Evidence 75/100How would you design a loss function that encourages recommendation diversity?ML Fundamentals · hard · Evidence 75/100Explain Transformer fundamentals for a multimodal role.GenAI & LLM · medium · Evidence 75/100Explain LoRA and why it reduces fine-tuning cost.GenAI & LLM · medium · Evidence 75/100Implement MinStack with O(1) minimum lookup.Python & DSA · medium · Evidence 75/100Implement single-head attention in PyTorch.ML Coding & PyTorch · hard · Evidence 75/100Explain how deep neural networks differ from shallow models.ML Fundamentals · easy · Evidence 75/100Implement KL divergence in Python.ML Coding & PyTorch · medium · Evidence 75/100Add a causal mask that hides future tokens.ML Coding & PyTorch · hard · Evidence 75/100What are the key design choices in a computer-vision model?ML Fundamentals · medium · Evidence 75/100Explain core NLP modeling choices.GenAI & LLM · medium · Evidence 75/100Generate valid word combinations from a word list.Python & DSA · medium · Evidence 75/100Implement multi-head attention in PyTorch.ML Coding & PyTorch · hard · Evidence 75/100Explain BERT and its pretraining objectives.GenAI & LLM · medium · Evidence 75/100Explain Transformers in an NLP context.GenAI & LLM · medium · Evidence 75/100Compare major LLM fine-tuning techniques.GenAI & LLM · hard · Evidence 75/100How can Transformers be used in recommendation systems?GenAI & LLM · hard · Evidence 75/100Describe a hard technical challenge from an ML project.Project & Behavioral · medium · Evidence 75/100How would you reason about distributed-systems trade-offs in an ML product?ML System Design · hard · Evidence 75/100When would you choose SGD instead of Adam?ML Fundamentals · medium · Evidence 75/100Explain an NLP project and its modeling choices.Project & Behavioral · medium · Evidence 75/100Explain BLIP-2 at a high level.GenAI & LLM · hard · Evidence 75/100Solve a backtracking problem and explain pruning.Python & DSA · medium · Evidence 75/100Explain the architecture of attention.GenAI & LLM · medium · Evidence 75/100Design Gun-Violence Video ModerationML System Design · hard · Evidence 66/100Search in Rotated Sorted ArrayPython & DSA · medium · Evidence 66/100Binary Tree Maximum Path SumPython & DSA · hard · Evidence 66/100Explain supervised fine-tuning for LLMs.GenAI & LLM · medium · Evidence 66/100How would you explain a complex ML decision to a non-technical manager?Project & Behavioral · medium · Evidence 41/100What mathematical assumptions underlie the model you chose?ML Fundamentals · hard · Evidence 41/100Why do K-D trees degrade in high-dimensional spaces?ML Fundamentals · hard · Evidence 41/100How would you derive the mathematics behind an ML model you used in production?ML Fundamentals · hard · Evidence 41/100How would you turn a research model into a production ML system?ML System Design · hard · Evidence 41/100How do K-D trees accelerate nearest-neighbor search?ML Fundamentals · hard · Evidence 41/100How would you reason about product sense for an ML ranking feature?ML System Design · medium · Evidence 41/100How would you test whether a modeling choice actually improved the product?ML Fundamentals · medium · Evidence 41/100How would you defend the modeling choices in one of your papers or projects?Project & Behavioral · hard · Evidence 41/100When would you choose approximate nearest-neighbor search over exact search?ML Fundamentals · medium · Evidence 41/100Design concurrent LLM servingGenAI & LLM · hard · Evidence 37/100Estimate Transformer FLOPsML Math · hard · Evidence 34/100SGD vs AdamML Fundamentals · medium · Evidence 34/100Transformer vs RNNGenAI & LLM · medium · Evidence 34/100Mini-batch Training TradeoffsML Math · medium · Evidence 34/100Design a Second-Pass Feed RankerML System Design · hard · Evidence 34/100Design Hashtag-to-Video RelevanceML System Design · hard · Evidence 34/100Attention Score From Q, K, VML Math · medium · Evidence 34/100Implement and Optimize a Custom Neural Network LayerML Coding & PyTorch · hard · Evidence 34/100Implement a Transformer BlockML Coding & PyTorch · hard · Evidence 34/100Learning Rate SchedulingML Fundamentals · medium · Evidence 34/100QKV in Multi-Head AttentionGenAI & LLM · medium · Evidence 34/100Code Multi-Head Self-AttentionML Coding & PyTorch · hard · Evidence 34/100Implement a Transformer encoder layerML Coding & PyTorch · medium · Evidence 31/100