This guide covers what a Data Scientist at University of Toronto is expected to do and how to prepare for the interview.
Technical Screening
reportedThis round decides whether someone can hand you a schema and a question and trust the number that comes back. Correctness under a clock is the bar, not clever syntax. The habit that separates strong from weak answers is checking the grain: after every join, know how many rows you expect and whether the count moved. Most wrong answers in this format are not wrong logic, they are a fan-out from a key that turned out not to be unique, or a filter applied before an aggregate when it belonged after. Say what you expect before you run it.
What to demonstrate
- Whether your row counts survive each join, and whether you notice on your own when they do not
- Deliberate handling of rows that fail to match, including whether the question needs an inner join or a left join with the non-matches kept and counted
- Whether NULLs are treated on purpose, given that a NULL compares equal to nothing and that COUNT of a column skips it
- Reaching a defensible answer inside the window instead of a refined one after it
How to prepare
- Take a two-table schema, write a join that fans out on purpose, then fix it by collapsing the many-side to one row per key before joining. Repeat until the fix is reflex rather than recall.
- Write a funnel as one query and print the distinct user count at each stage, then confirm each stage is a subset of the one above it rather than assuming it
- Do a few timed runs in a plain text box with no autocomplete and no formatter, since assessment editors often have neither
Case Study Rounds
reportedUnderneath the business framing, this round is usually asking whether you can turn a fuzzy goal into a quantity that could be computed from data such a business would plausibly hold. That means a metric with a stated numerator, denominator, eligibility rule and time window, plus an honest account of the conditions under which it would mislead you. Answers come apart when a candidate names a familiar metric and never defines it, because every follow-up then lands on an ambiguity that was left open and the candidate has to invent the definition under pressure.
What to demonstrate
- Whether a named metric arrives with its denominator, eligibility rule and window attached rather than assumed
- Whether the measure follows from the mechanism you proposed, or is a recognisable metric retrofitted to it afterwards
- Whether you name a guardrail that would reveal the gain came from somewhere you did not want it to come from
- Whether you can say what data the plan requires and what you would settle for if that logging were never implemented
How to prepare
- Take five metrics you reach for by reflex and write each as one sentence containing numerator, denominator, eligibility rule and time window. The ones you cannot finish are the ones that will fail under follow-up.
- For a product you use daily, write the measurement plan you would propose for a change to it: primary metric, one guardrail, the unit of analysis, and the table the numbers would come from.
- Practise the substitution question. For three metrics you like, write what you would measure instead if the event you depend on were not being logged.
Behavioral Competencies
reportedMost of the weight in this round sits on the disagreement questions. Data work routinely produces an answer someone senior did not want, and the interviewer is trying to learn what you do in that hour. Both failure modes are common: folding as soon as a director pushes back, and treating the pushback as ignorance to be corrected with a better chart. A strong answer usually contains a specific thing the other person knew that you did not, and describes how you found out whether it changed the conclusion.
What to demonstrate
- Whether you can state the other side's argument accurately before you explain why you disagreed
- What you treated as evidence during the disagreement, such as a rerun under their assumption or a holdout check, rather than persuasion technique
- Whether you distinguish being overruled from being wrong, and can give an example of each
How to prepare
- Write out one disagreement where you turned out to be wrong, and say what in the data misled you. Candidates prepare the story where they were right, and the follow-up asks for the other one.
- For your main disagreement story, be ready to say what result would have made you drop your position. If no such result exists, you were not arguing from the data.
- Practise stating the opposing position out loud in one sentence the stakeholder would accept, then continue the story.
PracHub editorial advice for the preparation topics above.
Roster sync creates and deactivates accounts in bulk, and it is not behaviour.
An automated roster import at term start can create thousands of accounts in one hour and deactivate thousands more at term end. Signup, activation, and churn series computed without filtering enrollment_source IN ('roster_sync','admin_bulk') will show spikes and cliffs driven entirely by an integration job. It also breaks cohort retention: a roster-created account that never activates is a provisioning artefact, not a churned learner.
Pre/post gain studies that select on low pretest scores manufacture improvement.
Any measure with reliability below 1 produces regression to the mean, so a group chosen for scoring in the bottom quartile will score higher on retest with no intervention at all. The apparent gain scales with measurement error, which for a short quiz is large. The fix is a control group selected by the identical rule, or a design that models the pretest as a covariate rather than as a selection filter.
Comparing periods without accounting for seasonality or day-of-week
Compare whole weeks against whole weeks and check whether the same swing appeared in prior cycles or prior years before attributing it to anything you changed. Weekday and weekend populations often differ enough that a Tuesday-to-Saturday comparison is meaningless.
Treating a non-significant result as proof of no effect
Say whether the confidence interval excludes the effect sizes you would have cared about. If it does not, the honest reading is that the test was underpowered, so report the minimum detectable effect the design could have found and what sample size would resolve it.
Choose a category, try a prompt, then open its approach, worked solution or follow-up when you need it.
Explain the difference between frequentist and Bayesian approaches in …
Explain the difference between frequentist and Bayesian approaches in the context of academic research.
Approach
- Say what the estimate is of, and over what population it generalises.
- Sanity-check the answer against a simple bound or a simulated case.
- Quantify uncertainty explicitly rather than reporting a point estimate alone.
Follow-up
- What sample size would you need to detect an effect half this size?
- How would you explain this result to someone who does not know statistics?
Bootstrap a clustered interval for minutes per mastery event
You have weekly with section_id, learner_id, active_minutes, and mastery_events for one ISO week. The guardrail metric is total active_minutes divided by total mastery_events across the population, a ratio of sums rather than a mean of per-learner ratios. Write a cluster bootstrap in numpy: resample sections with replacement, recompute the ratio, and return a 95 percent percentile interval from at least 2,000 replicates. Also return the interval you get resampling learners, and state which one you would report and why.
Approach
- Reduce to a section index first: per section, the total active_minutes and the total mastery_events. Because the statistic is a ratio of sums, a replicate needs only those two totals, which makes thousands of replicates cheap and keeps the resampling unit explicit.
- Resample section indices with replacement up to the original section count, sum both totals over the resampled sections, then divide. Never average per-section ratios: that is a different estimand and it is dominated by whichever sections happen to be small.
- Guard the denominator. A replicate drawing only sections with zero mastery events is undefined. Count how often that happens and report it rather than dropping those replicates silently, because a high rate means the weekly window is too short for this metric and the interval is not the real problem.
- Take the 2.5th and 97.5th percentiles of the replicate ratios. Percentile intervals are adequate here only while the replicate distribution is roughly symmetric; check that before reporting, and move to BCa if it is skewed.
- Repeat with learners as the resampling unit and compare widths. Report the section interval. The learner interval understates width by roughly the square root of the design effect, and it is the one that will be quoted back at you when the guardrail later moves.
Worked solution 30 min
- sec = weekly.groupby('section_id').agg(mins=('active_minutes','sum'), ev=('mastery_events','sum')); point = weekly.active_minutes.sum() / weekly.mastery_events.sum().
- idx = rng.integers(0, len(sec), size=(B, len(sec))); num = sec.mins.values[idx].sum(axis=1); den = sec.ev.values[idx].sum(axis=1).
- Drop or flag replicates where den == 0, count them, then ratios = num[ok] / den[ok].
- ci_section = np.percentile(ratios, [2.5, 97.5]); repeat the same procedure resampling learner rows to get ci_learner.
- Report point, both intervals, their widths, the width ratio, and the zero-denominator replicate count.
Follow-up
- Sections range from 8 to 60 learners. Does resampling sections with replacement bias the point estimate, the interval, or neither?
- How does the interval change if you resample learner-weeks over four weeks instead of one, and what new assumption does that introduce?
- The interval comes back wide enough to contain both a 10 percent improvement and a 10 percent regression. What do you tell the team to do?
Compute pace adherence at term-week six
Given fct_enrollment (enrollment_id, learner_id, course_id, section_id, term_id, enrollment_source, enrolled_at, scheduled_start_date, due_date, completed_at, units_total, units_completed, status) and dim_term (term_id, term_start_date, term_end_date, term_length_weeks), compute the share of enrollments at or ahead of pace at the end of term-week 6. An enrollment is on pace when units_completed >= units_total * (6 / term_length_weeks). Count only enrollments active at that moment. Handle NULL scheduled_start_date and units_total of zero explicitly, and return one value per term with its denominator.
Approach
- Join to dim_term on term_id and derive the cutoff as term_start_date plus 42 days. The metric is anchored to the term calendar, not to each enrollment's own start, which is what makes it comparable across institutions on different calendars.
- Resolve NULL scheduled_start_date by falling back to term_start_date, and count how often the fallback fired. A large fallback share is a finding about roster sync, not a detail to bury.
- Reconstruct active-at-cutoff rather than trusting the status column: enrolled_at <= cutoff and (completed_at is null or completed_at > cutoff). If status is current-state only, say so, because then filtering on it removes learners who fell behind and dropped later.
- Exclude units_total = 0 from both numerator and denominator and report that count separately. Left in, 0 >= 0 marks every empty enrollment as on pace and drags the rate up.
- Group by term_id and return numerator, denominator, and rate together, never the rate alone.
Follow-up
- A two-week holiday falls inside weeks one to six for one term. How do you stop that punishing those enrollments without hand-editing the threshold?
- Two terms have term_length_weeks of 10 and 16. Is a single week-6 number comparable across them, and what would you report instead?
Explain how you would use SQL window functions to calculate a rolling …
Explain how you would use SQL window functions to calculate a rolling average of research participation rates.
Approach
- Check whether any join is one-to-many before aggregating, or the sums inflate.
- Say which table is the grain you start from, and join outward from it.
- State the window function and its partition and ordering out loud before writing it.
Follow-up
- What breaks if events arrive late or out of order?
- How would you verify this result without re-running the same query?
How do you ensure data integrity when merging datasets from multiple a…
How do you ensure data integrity when merging datasets from multiple academic sources?
Approach
- Check whether any join is one-to-many before aggregating, or the sums inflate.
- State the window function and its partition and ordering out loud before writing it.
- Compute rates by summing numerator and denominator separately, never by averaging rates.
Follow-up
- How would you verify this result without re-running the same query?
- How does the query change if the join becomes one-to-many?
Rebuild sessions and active seconds from heartbeat events
Given raw_heartbeat(learner_id, content_item_id, content_version_no, heartbeat_at TIMESTAMP UTC, session_hint VARCHAR NULL), reconstruct the grain of fct_lesson_activity: one row per learner per content item per session. Close a session after 30 minutes with no heartbeat. Compute active_seconds as the sum of inter-heartbeat gaps inside a session with each gap capped at 120 seconds, and wall_seconds as last minus first heartbeat. Ignore session_hint, which the client sets unreliably. Then reconcile your active_seconds against the stored column in fct_lesson_activity and explain any systematic difference.
Approach
- Compute prev_at = LAG(heartbeat_at) OVER (PARTITION BY learner_id, content_item_id, content_version_no ORDER BY heartbeat_at) and gap_seconds from the difference. Partition on the version too, since the same item at a new version is different content.
- Flag is_new_session = (prev_at IS NULL OR gap_seconds > 1800), then session_seq = SUM(is_new_session::int) OVER (same partition ORDER BY heartbeat_at ROWS UNBOUNDED PRECEDING). The running sum over an ordered flag is the island assignment and it stays correct at any data volume, unlike a self-join on nearby timestamps.
- Aggregate per (partition, session_seq): active_seconds = SUM(CASE WHEN is_new_session THEN 0 ELSE LEAST(gap_seconds, 120) END), wall_seconds = MAX(heartbeat_at) - MIN(heartbeat_at), plus heartbeat_count.
- State the two consequences of the definition before anyone asks: a single-heartbeat session has active_seconds = 0 and wall_seconds = 0, and the per-gap cap means active_seconds can never exceed 120 * (heartbeat_count - 1), so a client with a slow heartbeat interval understates real time.
- Reconcile by joining on (learner_id, content_item_id, started_at) and examining the distribution of the difference, not a row count. A uniform shift points at the cap or the close rule; a heavy one-sided tail points at heartbeats the stored column excludes.
Worked solution 30 min
- Write the LAG and gap CTE and eyeball ten rows for one busy learner before adding anything else.
- Add the is_new_session flag and the running-sum session_seq, then verify session_seq increments exactly where the gap exceeds 1800 seconds.
- Aggregate to the session grain with the CASE-guarded capped sum.
- Join to fct_lesson_activity on learner, item and started_at, and plot or bucket the signed difference in active_seconds.
Follow-up
- The stored column excludes backgrounded tabs but raw_heartbeat has no visibility flag. How would you detect backgrounded stretches from timing alone, and how confident should you be?
- Two heartbeats can share a timestamp. What does that do to LAG, and does it change the answer?
- A learner studies for two hours with a 40-minute dinner break in the middle. Your rule gives two sessions. Which downstream metrics care about that choice and which do not?
How do you prioritize your time when balancing multiple research or in…
How do you prioritize your time when balancing multiple research or instructional projects?
Approach
- State what result would change your recommendation, so the answer is falsifiable.
- Decompose the metric into the rates that drive it, and say which one you would check first.
- Name one primary metric, then the guardrail that stops it being gamed.
Follow-up
- How would you detect that the metric is being gamed rather than genuinely improving?
- What would you do if the primary metric and the guardrail moved in opposite directions?
If you notice a sudden drop in user engagement with a research dashboa…
If you notice a sudden drop in user engagement with a research dashboard, how would you diagnose the root cause?
Approach
- Name one primary metric, then the guardrail that stops it being gamed.
- Decompose the metric into the rates that drive it, and say which one you would check first.
- Restate the decision this analysis has to support, and who acts on the answer.
Follow-up
- How would you detect that the metric is being gamed rather than genuinely improving?
- Which segment would you cut first, and what would that rule out?
How would you design a metric to evaluate the success of a new student…
How would you design a metric to evaluate the success of a new student-facing data platform?
Approach
- State what result would change your recommendation, so the answer is falsifiable.
- Fix the population and the time window before naming any metric.
- Restate the decision this analysis has to support, and who acts on the answer.
Follow-up
- Which segment would you cut first, and what would that rule out?
- How would you detect that the metric is being gamed rather than genuinely improving?
Define the key product metrics for a data-driven health application.
Define the key product metrics for a data-driven health application.
Approach
- Name one primary metric, then the guardrail that stops it being gamed.
- Decompose the metric into the rates that drive it, and say which one you would check first.
- Restate the decision this analysis has to support, and who acts on the answer.
Follow-up
- How would you detect that the metric is being gamed rather than genuinely improving?
- What would you do if the primary metric and the guardrail moved in opposite directions?
How do you determine the required sample size to reach statistical sig…
How do you determine the required sample size to reach statistical significance?
Approach
- Name the guardrails that would stop a launch even on a positive primary result.
- Decide the analysis before seeing data, including how long it runs and when you look.
- Say whether units interfere with each other, and switch design if they do.
Follow-up
- How would you handle interference between treated and control units?
- What would you do if you could not randomise at all?
If an A/B test shows a significant result but the effect size is negli…
If an A/B test shows a significant result but the effect size is negligible, how do you proceed?
Approach
- Name the guardrails that would stop a launch even on a positive primary result.
- Say whether units interfere with each other, and switch design if they do.
- Decide the analysis before seeing data, including how long it runs and when you look.
Follow-up
- What would you do if you could not randomise at all?
- What would you conclude if the result is positive but the test is underpowered?
Decide what an administrator usage report counts
You must publish a five-line usage report that institutional administrators read before renewal. Once it is published, instructors will assign work to move whatever it counts. Available: fct_lesson_activity (active_seconds, is_assigned, completion_status), fct_assessment_response (max_points, scoring_mode, is_correct, response_seconds, hint_count), fct_enrollment (units_completed, units_total) and dim_learner (first_activity_at, age_gated). Choose the five lines and their source columns, name one metric you refuse to publish and the behaviour it would cause, and state the pairing that keeps each line honest.
Approach
- Start from behaviour rather than from data availability. Publishing a number to the buyer converts it into an assignment, so the test for each line is whether its gamed version is still something you want happening.
- Choose lines that survive that test: seat activation with the never-activated count shown rather than hidden; learners with at least one scored submission in the last four weeks; pace adherence at the current term-week; verified mastery events with the reporting lag printed; and objectives with no coverage yet, which is a gap list, is directly actionable, and cannot be inflated.
- Name the source columns per line so the report is reproducible: first_activity_at against seats_provisioned, fct_assessment_response.submitted_at with scoring_mode, units_completed over units_total against the term dimension, the mastery ledger with its delayed check, and the objective coverage join.
- Refuse active minutes per learner. Publishing it causes seat-time assignments, which is the clearest case of teaching to the number in this domain and the weakest claim the product can defend as learning. Keep it internally as a guardrail where it is useful and harmless.
- Pair every countable line with an integrity partner: pace adherence with assessment integrity (response_seconds above the item floor, hint_count below cap), submissions with mean max_points so splitting one task into five cannot move the line, activation with the raw never-activated count.
- State the population on the report itself, including the age_gated share and any excluded enrollment sources, because those learners are missing from the behavioural tables and the report will otherwise look complete while describing a partial population.
Worked solution 30 min
- For each candidate line, write the one sentence describing how an instructor would move it with no learning, and drop the lines whose gamed version is undesirable.
- Write the five surviving lines with their source columns and windows.
- Attach an integrity or substance partner to each countable line, and write the single sentence justifying the refusal of active minutes per learner.
- Add the population footer: age_gated share, excluded enrollment sources, and the reporting lag on verified mastery.
Follow-up
- An administrator says your report shows fewer active learners than their own dashboard. What are the three likeliest causes?
- A district asks for a per-teacher leaderboard built from these lines. What do you do?
- If you could publish only four lines, which one goes, and what breaks?
Active minutes fell eighteen percent with session count flat
Weekly active learning minutes per active learner fell from 42 to 34 the week a new client build shipped. Distinct learner_id and row count in fct_lesson_activity are both flat week over week. You have fct_lesson_activity (session_id, started_at, ended_at, active_seconds, wall_seconds, device_type, content_version_no, activity_date) and nothing else today. Decide whether this is a real engagement change or an instrumentation change, name the column that settles it, and write the query. Deliverable: a one-paragraph call within the day.
Approach
- Compute the daily median of active_seconds / wall_seconds per session for fourteen days either side of the release. active_seconds is heartbeat-derived and capped per gap; wall_seconds is ended_at minus started_at and is untouched by heartbeat cadence, so the ratio isolates the instrument from the behaviour.
- Hold the denominator fixed alongside it: sessions per learner per week and median wall_seconds per session. If both are flat, nothing about how long learners stayed changed, which leaves only how time was counted.
- Split the ratio by device_type. Managed Chromebook fleets typically run a pinned build and update on a school IT cycle, so that group is a natural unexposed comparison for a client-side change.
- State the mechanism explicitly: active_seconds sums inter-heartbeat gaps with each gap capped at 120 seconds, so a heartbeat interval raised above 120 seconds discards the excess on every gap and shrinks active_seconds proportionally while wall_seconds is unaffected.
- Size it: apply the pre-release ratio to post-release wall_seconds to get counterfactual minutes, and report the difference from the published number as the artefact.
Follow-up
- What would you backfill, and what would you leave as a documented break in the series?
- If wall_seconds had fallen by the same proportion, what would you check next before calling the drop real?
- How would you make this metric robust to the next heartbeat cadence change without re-litigating it?
For someone who can already write the query and train the model but stalls when asked what to measure or whether a change is worth making. Metric definition and case structure come first; the technical work is kept as maintenance rather than the centre of the week.
Prepare, practise & reflect
One practical outcome each day. Spend longer where you need it.
0 / 7 done01Metric anatomy
- For three products you use daily, write one primary metric, two input metrics that plausibly move it, and one guardrail that would catch a cheap way of moving the primary at the cost of the product.
- For one of them, specify the metric precisely enough that two analysts would return the same number: numerator, denominator, unit of observation, time window, and how returning and deleted accounts are treated.
- Pick a ratio metric and write what happens to it when the denominator shrinks for reasons unrelated to the numerator, with a concrete example of that happening.
Deliverable: A one-page metric tree for one product, with the primary metric written as an unambiguous spec.
Practice prompt ↗Practice prompt ↗Practice prompt ↗Worked solution ↗02Diagnosing a drop without guessing
- Take the prompt "weekly active users fell 8 percent week over week" and write the segmentation plan before proposing any cause: platform, region, tenure cohort, acquisition channel, and whether the movement sits in the numerator or in a changed denominator.
- List the instrumentation failures that manufacture fake drops (a client release that stopped firing an event, a bot filter change, a shifted date boundary or timezone) and write the query that rules out each one.
- Rehearse stating the boring explanations first, seasonality and day-of-week composition, before reaching for a product cause.
Deliverable: A drop-diagnosis checklist short enough to recite from memory in under a minute.
Practice prompt ↗Practice prompt ↗Practice prompt ↗03Should we build it
- Take a feature idea and write it as a bet: what you believe is true, what would have to be true for it to pay off, the metric that would confirm it, and the effect size that would justify the engineering cost.
- Size the opportunity top-down and bottom-up, then reconcile the two numbers in writing instead of quoting whichever is friendlier.
- Write the counter-metric that would make you kill the feature even if it wins on the primary metric.
Deliverable: A one-page product memo ending in a decision rather than a list of considerations.
Practice prompt ↗Practice prompt ↗Practice prompt ↗04The places aggregate numbers lie
- Construct a Simpson's paradox numerically: two segments where the treatment wins within each segment yet loses overall, and identify the shift in segment weights that causes it.
- Take a heavy right-tailed quantity such as revenue per user and write why the mean is the wrong summary, which percentile you would report instead, and what a moving mean with a stable median tells you.
- Write your definition of a session for the product from day one, then name two real behaviours it misclassifies.
Deliverable: One page holding a worked Simpson's paradox table and a session definition with its two known failure cases.
Practice prompt ↗Practice prompt ↗Worked solution ↗05Technical maintenance, aimed at metrics
- Solve four timed SQL prompts that all end in a ratio metric, so the question of grain stays live in every answer.
- Compute a 95 percent confidence interval for a proportion on a small sample, and state why the normal approximation is unreliable when either np or n(1 minus p) falls below roughly 10, along with which interval you would use instead.
- Take one metric from your day-one tree, write the query that computes it correctly, then write the query that computes it wrong in the most plausible way and explain how you would notice.
Deliverable: Four solved prompts plus a matched correct and plausible-wrong query for one metric.
Practice prompt ↗Practice prompt ↗06Turning engineering work into data science stories
- Write three project stories as situation, decision, trade-off, outcome, each carrying one number and one thing you got wrong.
- For the story you will lead with, prepare an answer to "what would you do differently" that names a decision you made, not a constraint you were handed.
- Practise the sentence that reframes a systems project as a question project: the question the work answered, ahead of the pipeline it shipped.
Deliverable: Three written stories with the lead story delivered aloud and timed under four minutes.
Practice prompt ↗Practice prompt ↗07Mock case and gap list
- Run a 40-minute mock case with someone playing a product manager who pushes back on your metric choice, and record it.
- Listen back and mark every moment you proposed a solution before the success metric existed.
- Rewrite those moments as the question you should have asked, and rehearse the first 90 seconds of the case until scoping comes before solving.
Deliverable: A recorded case plus a rewritten opening 90 seconds.
Practice prompt ↗Practice prompt ↗Worked solution ↗Expand any day for tasks and deliverables. Your progress is saved on this device.
A number you shipped turned out to be wrong, and someone had already acted on it. That is one of the most useful stories a data person can carry. What is being scored is how fast you noticed, who you told first, and what you changed in the process so the same class of error could not repeat quietly.
Tell me about a time you had to explain a complex technical finding to…
Tell me about a time you had to explain a complex technical finding to a non-technical stakeholder.
Approach
- Quantify the outcome, including what you would not claim credit for.
- Pick a story where you drove the decision, not one where you observed it.
- Close with what you would do differently, concretely.
Follow-up
- How did you know the outcome was caused by your change?
- What would you do differently if you ran that project again?
Explain a risk score to a non-technical sales executive
Your renewal-risk score joins fct_subscription_period to seat activation from dim_learner.first_activity_at and to pace adherence from fct_enrollment. For one institution with 900 seats_provisioned and 31% activation at term-week 6, it outputs a 0.62 probability of non-renewal. A sales VP asks whether you are losing the account or not. You have two minutes and no slides. Deliverable: the spoken answer, plus the one line you would put in the weekly account review so the number is not later quoted as a certainty. Probed: whether you can carry uncertainty and still give a decision.
Approach
- Answer the question that was asked before qualifying it. 'Treat this as at-risk and act this week' is the answer; the probability is the support for it.
- Translate the score into a frequency on a reference class the VP already trusts, such as how many of the last fifty accounts scored in the 0.55-0.70 band did not renew, instead of explaining calibration in the abstract.
- Name the driver that is still movable before period_end: 31% seat activation at term-week 6 can be changed this term, whereas last term's usage cannot, so that is where the conversation goes.
- State what would move the score and by when, so the number arrives attached to an action rather than as a verdict.
- For the written line, use a band and a review date rather than a bare decimal, because a single number in a document will be re-quoted without its interval.
Follow-up
- The VP asks for every account scored above 0.50. What do you tell them about where that threshold came from?
- How would you check whether 0.62 is calibrated at all, and on what sample?
Account for an analysis you later discovered was wrong
Describe a case where you reported a result and later found it was wrong. Choose one where the error was yours and the number had already been used for a decision. Cover: the claim, the actual defect at the data or design level, how you found it, how long it had been live, who you told and in what order, and what you changed so the same class of defect cannot recur. The interviewer will push on the mechanism, not the apology. Probed: whether you understand your own failure modes well enough to have engineered around them.
Approach
- Pick an error with a nameable mechanism, such as a grain mistake, a survivorship filter, or a join that silently dropped age_gated learners. 'I misread a chart' has nothing testable in it.
- State the blast radius in decisions rather than in dashboards: what was decided, what was spent or promised, and to whom.
- Describe detection honestly, including whether someone else caught it and how long it ran undetected. A slow external catch is a usable answer; a vague one is not.
- Give the disclosure order and the reason for it: the decision owner first, with the corrected number and its direction in the opening sentence, then the wider audience.
- Close on the systemic fix and say whether it has fired since. A control that has never triggered on anything is a resolution, not a fix, and saying so is worth more than claiming otherwise.
Follow-up
- What did the corrected number change about the decision? If nothing changed, why did you escalate at all?
- Has the check you added caught anything since, and on what?
- What class of error would still get past you today?
- 01
Tell me about a time you had to explain a complex technical finding to a non-technical stakeholder.
- 02
Your renewal-risk score joins fct_subscription_period to seat activation from dim_learner.first_activity_at and to pace adherence from fct_enrollment. For one institution with 900 seats_provisioned and 31% activation at term-week 6, it outputs a 0.62 probability of non-renewal. A sales VP asks whether you are losing the account or not. You have two minutes and no slides. Deliverable: the spoken answer, plus the one line you would put in the weekly account review so the number is not later quoted as a certainty. Probed: whether you can carry uncertainty and still give a decision.
- 03
Describe a case where you reported a result and later found it was wrong. Choose one where the error was yours and the number had already been used for a decision. Cover: the claim, the actual defect at the data or design level, how you found it, how long it had been live, who you told and in what order, and what you changed so the same class of defect cannot recur. The interviewer will push on the mechanism, not the apology. Probed: whether you understand your own failure modes well enough to have engineered around them.
Is this an official University of Toronto interview guide?
No. It is PracHub's own research and practice material for the Data Scientist role at University of Toronto. Rounds and questions reflect what candidates have reported, not a process University of Toronto has published, and they change over time. Confirm the current format and scope with your recruiter.
PracHub interview research ↗How long should I expect the interview process to take?
The timeline can vary, but generally, it spans several weeks from the initial screening to the final decision. Be prepared for a thorough evaluation that respects the academic calendar.
PracHub interview research ↗What is the most common reason for a candidate not progressing?
A lack of structure in problem-solving is often a hurdle. Candidates who jump straight into technical solutions without defining the business or research problem often struggle to demonstrate the level of seniority required.
PracHub interview research ↗Will I need to do a live coding challenge?
Yes, expect technical rounds that involve SQL and data manipulation. Practice writing clean, readable code under time pressure to build your confidence.
PracHub interview research ↗How much focus is there on machine learning?
While the role is product and statistics-focused, having a solid understanding of machine learning basics is beneficial, particularly for predictive modeling tasks that support research.
PracHub interview research ↗Sources & methodology 3 sources ↗
No official company page is cited. Rounds and questions come from candidate reports and PracHub editorial material; each source shows the date it was read.
- 01PracHub interview research ↗
PracHub editorial research into this company and role, maintained with this guide. Candidate-reported, not an employer publication.
PracHub page · Accessed 2026-09-22 - 02PracHub Data Scientist practice ↗
Cross-company practice questions for this role.
PracHub page · Accessed 2026-09-22 - 03PracHub interview preparation framework ↗
The framework the preparation plan follows.
PracHub page · Accessed 2026-09-22
Explore related engineering environments and compare the preparation focus.
Yale University Data Scientist Interview Questions & Guide 2026
Timed practice for Yale University data scientist candidates who freeze under a clock: short SQL and probability sets run end to end, then reviewed against a written rubric.
Universitätsklinikum Tübingen Data Scientist Interview Questions & Guide 2026
A prep plan for Universitätsklinikum Tübingen data scientist candidates built on spaced review: write a full answer before opening any solution, keep a running list of the assumptions you failed to state, and repeat every missed exercise th
University of Michigan Data Scientist Interview Questions & Guide 2026
Metric-diagnosis practice for University of Michigan data scientist candidates: a number moves week over week and you work out why, splitting by segment, ruling out a logging or release change, and separating a mix shift from a real change
University of Florida Data Scientist Interview Questions & Guide 2026
Causal-inference exercises for University of Florida data scientist prep: difference-in-differences, matching, instrumental variables, and knowing when observational data simply cannot answer the question asked.