TD Data Scientist Interview Questions
The questions to prepare for a TD Data Scientist interview. Questions from real interview reports rank first. Updated daily.
Design an A/B test for a new platform feature, including success metrics, power, guardrails, and a clear ship decision.
Balance predictive performance with fairness checks and interpretable explanations when using complex black-box models.
Sign up to see every question
Create a free account to unlock this list and practice real interview questions.
Best practices for reproducible dataset and model versioning in shared ML pipelines.
Design a real-time fraud scoring system for card transactions with strict latency, delayed labels, and high availability requirements.
Explain what a p-value means, how it relates to statistical significance, and how to describe it clearly to non-technical stakeholders.
Explain which metrics matter for evaluating a churn model and how to choose them based on retention costs and business goals.
Outline the first checks to diagnose a sudden drop in a core product metric, starting with data quality, scope, and decomposition.
Tests SQL proficiency for real-world data extraction and transformation.
Identify the top 5% of clients by transaction count during the previous calendar quarter.
Use GROUP BY and HAVING to find duplicate patient records in a Johns Hopkins Medicine dataset.
Emerson
CareDx
SberbankCalculate each study's six-month rolling average of patient enrollment using PostgreSQL window functions.