TD Bank Data Scientist reports describe two to four interview rounds spanning fit, coding, resume and project discussion, behavioral questions, and communication with non-technical stakeholders.
$126K
Avg. Base Comp
$156K
Avg. Total Comp
2-4 rounds
Typical Rounds
2-4 weeks
Process Length
TD Bank Data Scientist candidates report interview processes ranging from two to four rounds. One report described two rounds: an initial coding segment with two to three easy-to-medium LeetCode questions and discussion of past data science projects, followed by technical and behavioral questions. Another described an HR fit conversation followed by a hiring-manager discussion of projects, data cleaning, imbalanced datasets, and metric choices. A newer candidate reported an initial in-depth screen and said three additional rounds were expected.
Build a clear project walkthrough that explains the business problem, your data preparation, modeling decisions, and evaluation choices. Reported project discussions covered resampling or class weighting for imbalanced data and the use of recall or AUC. Candidates also encountered questions on the bias-variance tradeoff and the difference between XGBoost and Random Forest.
Communication is prominent in the reports. Prepare examples of explaining technical methods or results to non-technical audiences, collaborating with stakeholders, handling disagreement, and responding when you make a mistake. Coding preparation should include breadth-first search and practical Python and SQL questions; one candidate reported these in a one-hour later-round interview. Fraud or risk-modeling practices appeared in an initial screen, so be ready to connect relevant experience to practical decisions.
Synthesized from 4 candidate reports by our editorial team.
Had an interview recently?
Share your experience. Unlock the full guide.
Real interview reports from people who went through the Td bank process.
1 round (3 more to go), seemed like a good start and in depth screen. some basic scanning of my resume and had to answer simple behavioral questions. very clear outline of the process and stuff like that. I had to describe what I would do if I sent out a mistake
Questions asked: Explain a time you had to explain technical methods to a non-technical audience. Explain risk modeling or fraud detection practices. Bias/variance tradeoff.
Share your own interview experience to unlock all reports, or subscribe for full access.
Sourced from candidate reports and verified by our team.
Topics based on recent interview experiences.
Featured question at Td bank
Explain the difference between XGBoost and random forest and give an example where you would use one over the other
| Question | |
|---|---|
| Credit Card Fraud Model | |
| Client Solution Pushback | |
| Duplicate Product Names | |
| Uber Eats Success | |
| Justify a Neural Network | |
| Your Strengths and Weaknesses | |
| Interest Rates | |
| 2nd Highest Salary | |
| Empty Neighborhoods | |
| Rolling Bank Transactions | |
| Employee Salaries | |
| Merge Sorted Lists | |
| Subscription Overlap | |
| Comments Histogram | |
| Closest SAT Scores | |
| Top Three Salaries | |
| Cumulative Distribution | |
| Monthly Customer Report | |
| Experiment Validity | |
| Compute Deviation | |
| Find the Missing Number | |
| Bagging vs Boosting | |
| String Shift | |
| Button AB Test | |
| 500 Cards | |
| Prime to N | |
| Session Difference | |
| Top 5 Turnover Risk | |
| Like Tracker |
Synthesized from candidate reports. Individual experiences may vary.
Reports describe an early HR or screening conversation that may review the resume, motivation for the role, and communication style. One candidate was asked how they would explain technical methods to a non-technical audience and what they would do after sending out a mistake. That candidate described the screen as the first of four expected rounds.
One candidate reported two to three easy-to-medium LeetCode questions in an initial coding portion, including breadth-first search. Another said a later one-hour interview included coding, with about 20 minutes each of Python and SQL questions. Review core problem solving and be prepared to explain your approach clearly.
Hiring-manager and technical discussions reportedly examined prior data science projects in detail. Candidates discussed cleaning and manipulating data, addressing imbalanced datasets through resampling or class weighting, and selecting metrics such as recall or AUC. Other reported prompts included the bias-variance tradeoff and XGBoost versus Random Forest.
Behavioral questions appeared in both early and later interviews. Prepare specific examples of communicating results to non-technical clients, working with cross-functional partners, resolving disagreement, and making sound decisions after an error. One screen also asked about fraud detection or risk-modeling practices.
