Katkat

Company

Tower interview questions

3 practice questions in the style of Tower's interviews, covering ML System Design, Machine Learning.

Questions3

0 easy1 medium2 hard

Your progress0/3

Tower interview questions by the numbers

Last updated

Katkat has 3 Tower interview questions. The largest topics are Machine Learning (67%) and ML System Design (33%). By difficulty: 33% medium and 67% hard. The most common round is Technical Screen (100%). Frequent concepts include Backpropagation, Deep Learning, distributed-training and gpu-optimization.

Topics

Difficulty

  • Medium33% · 1
  • Hard67% · 2

Interview rounds

  • Technical Screen100% · 3

Roles

  • Data Scientist67% · 2
  • Machine Learning Engineer33% · 1

Frequent concepts

  • Backpropagation33% · 1
  • Deep Learning33% · 1
  • distributed-training33% · 1
  • gpu-optimization33% · 1
  • infrastructure33% · 1
  • Initialization33% · 1

Showing 1–3 of 3 results

  • Debug Oscillating GPU Utilization During TrainingHard

    Design a systematic debugging and architectural remediation strategy for a distributed deep learning training cluster exhibiting severe GPU utilization oscillation between 90% and 20%.

    TowerData ScientistML System DesignTechnical Screen
  • Explain Why Constant Neural-Network Initialization FailsHard

    This question tests the conceptual understanding of weight initialization strategies and the symmetry-breaking property required for neural network gradient-based optimization.

    TowerData ScientistMachine LearningTechnical Screen
  • Explain the KV Cache and Prefill-Decode Disaggregation in LLM ServingMedium

    Explain how the KV cache speeds up autoregressive decoding in transformer language models, estimate how much GPU memory it needs, and describe prefill-decode disaggregation in LLM serving. It tests understanding of compute-bound versus memory-bound inference phases, cache sizing and the trade-offs of splitting the phases across GPU pools.

    TowerMachine Learning EngineerMachine LearningTechnical Screen

Tower interview experiences

All 1 →
  • ML Infra Questions, Not a MatchTower · Machine Learning Engineer · Technical Screen · Jul 2026
    Rejected

    After getting rejected at the QR stage earlier, I got picked up by another MLE team. Through this I also learned a bit about how some quant firms work: different teams (pods) seem to hire independently, and the teams share one company-wide infra while their strategies are isolated from each other. The title for this role was MLE, but it actually sounded more like they needed someone for ML infra…

More companies