Interview guides · Vetegrity

Vetegrity Data Scientist Interview Questions & Guide 2026

A two-week prep plan for Vetegrity data scientist candidates: write your answer before reading the solution, revisit misses three days later.

The 60-second brief

The Data Scientist role at Vetegrity is central to the company’s mission of providing high-impact analytics, automation, and AI/ML solutions to the Intelligence Community and the Department of Defense. You will not be working on abstract problems; instead, you will be embedded in mission-driven workflows where your technical output directly influences operational efficiency and decision-making. The role demands a unique combination of engineering rigor and scientific curiosity, requiring you to navigate complex, often unstructured data environments to deliver actionable insights.

If the team owns experimentation, expect depth past a two-sample test: minimum detectable effect and its roughly inverse-square-root dependence on sample size (holding power, significance level and variance fixed), variance reduction from pre-period covariates, interference between units, and when a sequential design is the right call.

Focus areas

Turn a vague request into a measurable question
Size an experiment before anyone launches it
Separate novelty effects from durable behaviour change

Overview

What makes this role particularly compelling is the blend of autonomy and high-stakes responsibility. Whether you are building agentic AI frameworks, optimizing data ingestion pipelines, or developing interactive dashboards for mission stakeholders, your work bridges the gap between raw information and strategic intelligence. You will be expected to demonstrate "grit" and integrity, ensuring that the solutions you deploy are not only technically sound but also compliant and reliable in high-pressure, networked environments.

The Vetegrity interview process

Vetegrity candidates report 5 rounds · ≈ 4-6 weeks. The stages below are what candidates describe, not a published process.

Initial Technical Screen

This 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

Problem-Solving Discussion

An added round often puts you in front of someone outside the core hiring team: a partner engineer, a product owner, a domain expert, sometimes a more senior manager. The question they are really asking is not whether you can do the work but whether they would trust a number that came from you. That changes what a good answer looks like. Lead with what the decision cost and what it changed, keep the method available but not central, and be plain about the limits of your evidence. Overstating a result is the fastest way to lose this round.

What to demonstrate

Whether you can explain a technical choice to someone who will never read your code, without either flattening it into nothing or hiding inside jargon
Honesty about evidence strength: what the analysis establishes, what it only suggests, and what it cannot say at all
How you take disagreement, specifically whether you update on a good objection, hold your position with reasons, or fold on contact

How to prepare

Write the two-sentence version of your most technical project for a non-specialist, then check that neither sentence needs a method name to make sense.
For one result you are proud of, write the strongest objection someone could raise and a response that concedes the part of it that is correct.
Prepare one decision that turned out to be wrong: how you found out, what it cost, and what you changed afterwards. A senior cross-functional interviewer asks for this more often than a technical one does.

Cultural Fit Interview

An extra round usually exists because something is still open after the standard loop: a skill the earlier interviews did not sample, a level decision, or two interviewers who disagreed. It is rarely a rerun of what you already did well. Ask the recruiter who you are meeting, what function they sit in, and how long the session runs. That is an ordinary scheduling question, and the answer changes what you should prepare. What separates a strong candidate here is treating the round as a fresh evaluation with its own bar, rather than assuming earlier performance carries you through or sinks you.

What to demonstrate

Whether you can answer well on ground the earlier rounds did not cover, without leaning on what you already said to someone else
Consistency of the facts in your stories: the same sample size, timeframe, team size and scope of your own role as in earlier conversations
How you handle an unfamiliar format live, including whether you ask what kind of answer is wanted before producing one

How to prepare

Ask the recruiter for the interviewer's function, the length, and whether to expect a coding surface, a discussion, or a presentation. Preparing for a 30 minute conversation with a partner team is not the same work as preparing for a 60 minute technical block.
Write out what each earlier round actually covered, then list the two or three areas nobody probed. That gap is the most likely subject of the extra round.
Re-read the numbers in the project stories you have already told, so a second telling does not quietly contradict the first.

Technical Assessments

A handful of shapes account for most of what gets asked in this format: a ranking or deduplication inside groups, a running or rolling total, a period-over-period comparison, and a cohort tracked forward over time. Recognising the shape quickly is most of the speed here; deriving it from scratch while a clock runs is where the time goes. Know that a window function keeps every row while a GROUP BY collapses them, and know which one the question needs. If the exercise is in Python instead of SQL, the same shapes arrive as groupby with transform, shift and merge, and the same grain mistakes are available.

What to demonstrate

Whether you reach the right construct without a detour, such as ROW_NUMBER over a partition to deduplicate instead of a self-join against a MAX subquery
Whether you know what your window frame actually is, since adding ORDER BY inside OVER changes the default frame and silently changes a running total
Whether the thing runs. A near-miss that throws an error scores below a plainer query that returns the right rows.

How to prepare

Write each of the four shapes once from memory against a small schema and keep the working version somewhere you will reread it: dedupe with ROW_NUMBER, a running total, a month-over-month change with LAG, and a retention table
Compute one running total twice on data with tied timestamps, once on the default frame and once with ROWS BETWEEN UNBOUNDED PRECEDING AND CURRENT ROW, and look at where the two disagree
If Python is on the table, rebuild the dedupe and the running total with groupby and cumsum, then assert the two implementations return identical rows

Final Evaluation

A loop is not scored one interview at a time. The people you meet compare notes afterwards, usually in a meeting you are not in, and the outcome turns on what each of them can say about you when asked. That rewards something other than survival: every room needs one specific thing worth repeating, and none of them can contradict another. The common way to lose is to tell the same project four times with different numbers in it, or to be uniformly fine in a way that leaves nobody with anything to argue for.

What to demonstrate

Whether your account of a project survives being told twice, with the same scale, the same metric definition and the same numbers each time
Whether each interviewer leaves with one concrete claim they could make on your behalf later, rather than an absence of complaints
Whether a question you already answered in an earlier room gets the same answer at the same depth, without visible impatience

How to prepare

Write a one-page fact sheet for your two or three main projects that fixes the numbers you will quote: rows of data, the metric as a single sentence, the effect you measured and how long the work took. Say them aloud from the sheet until they come out identical every time
For each kind of room you expect, decide the one sentence you want that interviewer repeating in a debrief, then check during the mock that you said it outright instead of implying it
Rehearse answering the same project question twice in one sitting, the second time as though you had not just answered it, because the thing that needs fixing is the flatness that creeps into a repeated story

Where candidates lose points

Treating last-touch attribution as the causal value of a channel

The attribution label on dim_user is the output of a rule that assigns full credit to whichever touch happened to be recorded last inside a lookback window, and that rule systematically rewards channels that sit close to the conversion, especially branded search and retargeting, which largely intercept demand that already existed. Reallocating spend on those labels moves budget toward the channels that are best at being last, which is why attributed return on ad spend often improves while total signups do not. Nothing in the touchpoint data can settle this, because the counterfactual of not running the channel was never observed. The credible reads are a geo holdout or a scheduled pause, sized in advance on the total-signups metric rather than on the attributed one, and the honest framing in the meantime is that the label describes correlation with conversion and not incremental contribution.

Counting on an identity key that changes underneath the metric

visitor_id is per browser and per device, and it resets on cookie clearance, private browsing and platform privacy changes, so the distinct-visitor count drifts upward for reasons unrelated to reach. Any rate with visitors in the denominator therefore decays over time even when behaviour is constant, and any rate with visitors in the numerator inflates. The stitching at signup makes it worse in both directions: a user who signed up on mobile and returns on desktop is two visitors and one user, while a shared device is one visitor and several users. Decide which key each metric is counted on, write it into the definition, and when comparing a period before and after a platform privacy change, expect a level shift in every visitor-keyed metric and do not attribute it to the product.

Building features from data that postdates the prediction time

Check every feature against the timestamp at which the model would actually score, and drop anything computed from a window that includes or follows the label event. For a forecasting use case, split train and test by time rather than at random, and split by entity when the same entity recurs.

Reporting a p-value with no effect size or interval

Give the estimated difference with a confidence interval in the units the business cares about, then say whether that whole interval is worth acting on. A p-value only addresses whether you can rule out exactly zero; it says nothing about magnitude.

Questions & practice

What trade-offs do you consider when choosing between a high-recall and high-precision model for a mission-critical application?

Medium · machine learning and modelling

Approach

Pick an evaluation metric that matches the cost of each error type, not a default.
Set a baseline first, so any model has something honest to beat.
Say how the offline result would be validated online before it is trusted.

Follow-up

How would you choose the decision threshold, and who owns that choice?
Where could label leakage enter this setup?

Cluster bootstrap for a per-session rate randomised on users

Hard · cluster bootstrap · randomisation unit · variance

An experiment randomised on user_id reports a per-session conversion rate, so sessions inside a user are correlated. Input: one row per session with user_id, variant in {control, treatment} and converted in {0,1}. Write a cluster bootstrap from scratch: resample users with replacement within each arm, keep every session of a drawn user, recompute each arm's ratio of converted sessions to sessions, and take the difference. Return the point estimate, a 95 percent percentile interval from at least 2,000 resamples, the naive session-level interval that ignores clustering, and the ratio of their widths.

Approach

Name the estimand precisely: it is a ratio of sums, sum(converted) over sum(sessions) within an arm, not the mean of per-user rates. Those differ whenever session counts vary across users, and the ratio is what the reported metric is.
Resample the cluster, not the row. Draw n_users user ids with replacement inside each arm and take every session belonging to each draw, including duplicate draws of the same user. Keeping the user count fixed per arm rather than the session count is what preserves the sampling design.
Precompute per-user (converted_sum, session_count) once, so each resample is two vector lookups and a division rather than a repeated filter over the session frame. That turns 2,000 resamples from minutes into under a second.
Take the 2.5th and 97.5th percentiles of the 2,000 differences for the interval, and report the point estimate from the full data rather than from the bootstrap mean, since the bootstrap mean carries the resampling bias.
Compute the naive interval from the session-level binomial standard error and compare widths. The expected inflation is roughly sqrt(1 + (m-1)*rho), with m the mean sessions per user and rho the intraclass correlation of converted within users, so a computed ratio far from that value points at a bug in one of the two intervals.

Worked solution (35 min)

per_user = df.groupby(['variant','user_id'])['converted'].agg(['sum','size']); split into two arrays per arm.
point = (t_sum.sum() / t_n.sum()) - (c_sum.sum() / c_n.sum()).
For b in range(B): idx = rng.integers(0, len(t_sum), len(t_sum)); ratio_t = t_sum[idx].sum() / t_n[idx].sum(); same for control; store the difference. Vectorise by drawing a (B, n) index matrix if memory allows.
ci = np.percentile(diffs, [2.5, 97.5]); naive_se = sqrt(p_t*(1-p_t)/n_sessions_t + p_c*(1-p_c)/n_sessions_c); naive_ci = point +/- 1.96*naive_se.
width_ratio = (ci[1]-ci[0]) / (naive_ci[1]-naive_ci[0]).

Expected result: A point estimate identical to the full-data ratio difference, a percentile interval containing that point estimate, and a width ratio greater than 1 whose value approximates sqrt(1 + (m-1)*rho) for the data's mean cluster size m and intraclass correlation rho. A ratio of approximately 1.0 means sessions were resampled instead of users.

Verify it yourself

Construct data where every user has exactly one session: the cluster and naive intervals then agree to within Monte Carlo error, since m = 1 makes the design effect 1.
Construct data where converted is constant within each user: rho is 1, and the width ratio should approach sqrt(m).
The bootstrap distribution's mean sits close to the point estimate; a large offset means the resampled arms have drifted in size.
Doubling B narrows the Monte Carlo noise in the percentile endpoints without shifting the interval's centre.

Follow-up

Users average 3.4 sessions and the intraclass correlation is 0.12. What width ratio do you predict before running it, and does your bootstrap land there?
Give the delta-method standard error for this ratio and say when you would prefer it to the bootstrap.
Half the users in the treatment arm have exactly one session. What does that do to the cluster bootstrap's coverage, and how would you check it?

Rebuild per-visitor ordering without groupby convenience methods

Easy · pandas · vectorisation · window logic

You have a DataFrame of 2 million fct_event rows with visitor_id, occurred_at_utc and event_id, unsorted and containing duplicate timestamps within a visitor. Produce three new columns: event_rank, the 1-based position of the event within its visitor ordered by occurred_at_utc; seconds_since_prev, the gap to that visitor's previous event, NULL for the first; and is_first_for_visitor. You may use sort_values, shift, cumsum, numpy and boolean masking. You may not use groupby.transform, groupby.apply, groupby.cumcount, groupby.rank or merge_asof. Break timestamp ties on event_id.

Approach

Sort once by ['visitor_id', 'occurred_at_utc', 'event_id'] and reset the index. The whole exercise reduces to row arithmetic on a sorted frame, and the tiebreak on event_id is what makes the result reproducible across runs.
Mark visitor boundaries with is_first = df['visitor_id'].ne(df['visitor_id'].shift()). This is the single fact every other column derives from.
Compute seconds_since_prev as the diff of the timestamp column, then overwrite it with NaT/NaN wherever is_first is True. The shift crosses the boundary between visitors and will otherwise hand the first row of each visitor the last event of the previous one.
Build event_rank from a running counter that resets at boundaries: take a global cumulative position (np.arange(len(df))) and subtract, per row, the global position at which that visitor started. Get the start position by forward-filling the positions where is_first is True, which is a cumsum-free reset and is O(n).
Verify against the forbidden method once, as a test rather than as the implementation, and confirm the two agree on every row.

Follow-up

The frame does not fit in memory. How does your approach change if you can only process one visitor-partitioned chunk at a time?
occurred_at_utc is client-supplied and sometimes runs backwards within a visitor. Does your seconds_since_prev go negative, and should it?
How would you extend this to reset the counter at every change of surface as well as visitor?

Describe how you would use SQL window functions to calculate rolling averages or identify rank-based trends in a time-series dataset.

Medium · sql and data manipulation

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.
Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.

Follow-up

How does the query change if the join becomes one-to-many?
How would you verify this result without re-running the same query?

Explain how you would optimize a slow-running query that processes millions of rows of mission data.

Medium · sql and data manipulation

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

How does the query change if the join becomes one-to-many?
What breaks if events arrive late or out of order?

Pair flow starts to completions within a thirty-minute bound

Medium · event-streams · conditional-aggregation · flow-funnel

fct_event holds event_id, flow_id, flow_instance_id, event_name, occurred_at_utc, surface, app_version. Using only rows where event_name is 'flow_started', 'flow_completed' or 'error_shown', compute the daily core-flow completion rate by surface and app_version. Numerator: distinct flow_instance_id whose flow_completed occurs within 30 minutes of its flow_started with no error_shown for the same instance in between. Denominator: distinct flow_instance_id with a flow_started that day. Report rows with a NULL flow_instance_id as a separate coverage count, not inside either side.

Approach

Collapse the stream to one row per flow_instance_id using conditional aggregation: MIN(occurred_at_utc) FILTER (WHERE event_name = 'flow_started') AS started, MIN(...) FILTER (WHERE event_name = 'flow_completed') AS completed, MIN(...) FILTER (WHERE event_name = 'error_shown') AS first_error. This makes the 'first' semantics explicit and avoids a self-join entirely.
Apply the predicates on that single row: completed IS NOT NULL AND completed <= started + interval '30 minutes' AND (first_error IS NULL OR first_error > completed). Note the last clause encodes 'in between' literally, so an error surfaced after a successful completion does not disqualify the attempt.
Bucket by started::date, surface and app_version. Splitting by app_version is not decoration: a completion-rate regression is nearly always confined to one client build, and a pooled daily rate hides it behind the installed base.
Count NULL flow_instance_id rows as a separate coverage figure. They cannot be attributed to an attempt at all, so they belong in neither numerator nor denominator, and their share tells you how much of the rate is unmeasurable.
For any roll-up to week or to all surfaces, re-sum the numerator and denominator rather than averaging the daily rates.

Worked solution (25 min)

Write the per-instance collapse CTE and confirm COUNT(*) equals COUNT(DISTINCT flow_instance_id).
Add the three predicates one at a time, recording how many instances each removes.
Aggregate by day, surface and app_version with FILTER on the clean-completion flag.
Compute the NULL-instance coverage count as a separate select and report it alongside.
Pick the worst day-surface-version cell and pull ten raw instances to confirm the disqualification reason.

Expected result: One row per (day, surface, app_version) with started, completed_clean and a rate bounded by 0 and 1, plus a scalar coverage count of events carrying a NULL flow_instance_id. Instances completing after 30 minutes appear in the denominator only.

Verify it yourself

No rate exceeds 1.
COUNT(*) equals COUNT(DISTINCT flow_instance_id) in the collapsed CTE.
Total started across cells equals the count of distinct flow_instance_id values carrying a flow_started event in the window.
Instances with an error strictly between start and completion are excluded from the numerator but present in the denominator.

Follow-up

occurred_at_utc is client-supplied. What do you do with an instance whose completion timestamp precedes its start?
flow_instance_id is missing on 4 percent of starts for one app_version. What can and cannot you conclude about that build?
An instance starts at 23:55 and completes at 00:04. Which day owns it, and what does the alternative do to the daily series?

How would you design a dashboard to track the performance of an automated data ingestion pipeline?

Medium · product sense and metrics

Approach

Decompose the metric into the rates that drive it, and say which one you would check first.
State what result would change your recommendation, so the answer is falsifiable.
Name one primary metric, then the guardrail that stops it being gamed.

Follow-up

What would you do if the primary metric and the guardrail moved in opposite directions?
Which segment would you cut first, and what would that rule out?

How do you define "success" for an agentic AI tool designed to assist with intelligence gathering?

Medium · product sense and metrics

Approach

State what result would change your recommendation, so the answer is falsifiable.
Restate the decision this analysis has to support, and who acts on the answer.
Fix the population and the time window before naming any metric.

Follow-up

What would you do if the primary metric and the guardrail moved in opposite directions?
Which segment would you cut first, and what would that rule out?

If you notice a sudden drop in a key product metric, what is your systematic approach to diagnosing the root cause?

Medium · product sense and metrics

Approach

Decompose the metric into the rates that drive it, and say which one you would check first.
State what result would change your recommendation, so the answer is falsifiable.
Name one primary metric, then the guardrail that stops it being gamed.

Follow-up

What would you do if the primary metric and the guardrail moved in opposite directions?
How would you detect that the metric is being gamed rather than genuinely improving?

What are the most common experimentation pitfalls you have encountered, and how do you mitigate them?

Medium · a/b testing and experimentation

Approach

Say whether units interfere with each other, and switch design if they do.
State the primary metric and the minimum effect worth shipping, then size the test.
Name the randomisation unit first; it decides the variance and what the test can detect.

Follow-up

How would you handle interference between treated and control units?
What would you conclude if the result is positive but the test is underpowered?

How do you determine the required sample size for an A/B test to ensure statistical significance?

Medium · a/b testing and experimentation

Approach

Name the randomisation unit first; it decides the variance and what the test can detect.
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

What would you conclude if the result is positive but the test is underpowered?
What would you do if you could not randomise at all?

Reminder volume where the guardrail opposes the primary metric

Medium · engagement · guardrails · trade-offs

The growth team proposes tripling weekly reminder email volume. The north star is weekly active accounts completing a core action, counted on account_id from fct_event (is_core_action, account_id, occurred_at_utc, surface). Its stated guardrail is week-over-week repeat rate together with notification opt-out and unsubscribe rates. Reminders will move the primary up and the guardrail down, by design, and both effects are real. Define the decision rule before the test runs: what magnitudes make this a ship and what makes it a stop. Deliverable: the rule, including the exchange rate you are using between the two quantities.

Approach

Name the conflict precisely rather than calling it a balance: the reminder buys one week of an account returning and spends permission to contact that account, and permission is not renewable, so the two quantities are not comparable as percentage points.
Put both sides into one unit before arguing about thresholds. Value the primary gain as incremental core-action weeks over the horizon; value an opt-out as the forgone email-driven active weeks over that account's remaining expected lifetime.
Read opt-out as a stock, not a flow: accumulate it over the test, because a weekly opt-out rate that looks small is a cumulative curve that only ever rises within a fixed set of contactable accounts.
Run long enough for the novelty to decay and use the late number in the trade: compare week-1 lift with week-4 lift, state the decay you observed, and refuse to price the decision on a week-1 read.
Write the outcome as two numbers and a default action, including what happens when the result lands between them — hold and test a smaller volume increment rather than shipping on ambiguity.

Worked solution (30 min)

Measure the starting stock: current opt-out and unsubscribe rate per 1,000 contactable accounts, and the share of weekly active accounts whose session began from referrer_channel = 'email'.
Estimate the two effects separately over four weeks: incremental weekly active accounts per 1,000 extra sends, and incremental cumulative opt-outs per 1,000 extra sends.
Convert opt-outs into forgone active weeks: email-driven active weeks per account per year multiplied by remaining expected account lifetime, and set that against the incremental active weeks bought.
Compare week-1 with week-4 lift, state the decay rate, and carry the week-4 figure into the trade.
Write the rule as two thresholds plus a default action for the middle case.

Expected result: A written ship/stop rule containing a numeric exchange rate between one incremental weekly active account and one opt-out, derived from an observed lifetime and an observed decay rather than asserted.

Verify it yourself

The cumulative opt-out curve within a fixed set of contactable accounts must be non-decreasing. A dip means the denominator is being recomputed each week and the metric is not a stock.
Week-4 incremental lift should be materially below week-1. If it is not, either the horizon is too short to see decay or the arms are contaminated by cross-sends.
Repeat rate must be computed on the same accounts across consecutive weeks, not on the weekly active set, whose composition changes every week by construction.

Follow-up

Opt-out is flat but week-over-week repeat rate falls. What is the most likely mechanism, and does it change the decision?
You do not have a year of data to estimate remaining account lifetime. What do you substitute, and how do you keep the answer honest about that?
Does randomising on account rather than on user change either the readout or the size of the test?

Attribute a flow-completion drop to one client build

Medium · release-regression · instrumentation · staged-rollout

Core-flow completion rate on surface = 'ios', defined as distinct flow_instance_id reaching 'flow_completed' within 30 minutes of 'flow_started' with no 'error_shown' on the same id in between, fell from 78% to 71% over four days. A new iOS build began a staged rollout on day one. Using fct_event columns flow_id, flow_instance_id, app_version, event_name, user_id and occurred_at_utc, produce the completion rate per build per day, size the loss in absolute completions, and say whether the fix is a rollback. Treat the possibility that the instrumentation changed rather than the flow.

Approach

Cut the rate by app_version and day, each build against its own denominator. A pooled series during a staged rollout is a weighted average whose weights move daily, so it declines in proportion to the rollout share even when the old build is perfectly flat; confirming that proportionality is itself the evidence the regression is build-specific rather than environmental.
Test the denominator before believing the rate. Compute flow_started events per distinct user_id on each build: if the new build re-mints flow_instance_id on retry, or fails to carry the id from start to completion, the denominator inflates and the numerator falls with nothing changing for the user.
Normalise to a user-level outcome that is immune to the id question, namely completions per distinct user_id per day on each build. This is the number that says whether anyone actually failed to finish.
Check the error path: pull error_shown counts and their properties for flow_instance_id values on the new build. A genuine regression produces errors; an instrumentation break produces missing completions with no corresponding error volume, and the two prescribe different fixes.
Size the real loss as the per-user completion gap multiplied by daily users on the new build, and give the rollback recommendation conditional on the error evidence rather than on the rate.

Follow-up

The rollout is at 45% and product wants to go to 100% tomorrow. What do you say, and what would you need by when?
If the flow_instance_id is genuinely being re-minted, what is the correct historical treatment of the four affected days?
How would you have caught this on day one instead of day four?

Suggested practice checklist

For a candidate whose interviews will centre on A/B testing, metric movement and causal claims. Design comes before arithmetic, arithmetic before analysis, and the week ends by rehearsing the readout rather than the derivation.

Day 1 — Design one test end to end on paper

Take a single feature change and write the full design: randomization unit, the exact point of exposure, the primary metric with its grain, guardrails, allocation, planned duration, and the decision rule committed before any data exists.
Write why the randomization unit must sit at or above the level where treatment can spill over, and give one case where user-level randomization is still contaminated (shared accounts or devices, or two participants in the same marketplace).
State in advance what you will do if the primary metric is flat while a secondary metric is significant.

Deliverable: A one-page test design with a decision rule written before launch.

Day 2 — Power arithmetic until it is automatic

Compute required sample size per arm for a binary metric with the normal approximation, n is approximately 2 times (z for alpha/2 plus z for power) squared times p(1 minus p) divided by delta squared, for baselines of 2, 10 and 40 percent at a 5 percent relative lift, and note that for a fixed relative lift the requirement falls as the baseline rises because delta grows proportionally with p.
Redo the calculation for a continuous metric using variance in place of p(1 minus p), and show why a heavy-tailed quantity such as revenue per user needs either far more traffic or a capped version with a stated cap.
Convert one of the results into weeks given a weekly eligible traffic figure, then list the two honest ways to shorten it (accept a larger detectable effect, or reduce variance) and write why quietly lowering the power target is a decision to miss more real wins, not a speedup.

Deliverable: A small script or sheet that maps baseline, minimum detectable effect, alpha and power to sample size and weeks, cross-checked against a published calculator.

Day 3 — Variance and the unit-of-analysis problem

Take a ratio metric whose denominator is not the randomization unit (clicks per session, randomized by user) and compute the standard error twice, once naively at session level and once by the delta method or a user-level bootstrap, then record how much the naive version understates it.
Implement CUPED on simulated data: choose a pre-period covariate X measured before assignment, estimate theta as Cov(Y, X) divided by Var(X), and analyse Y minus theta times (X minus its mean) in place of Y. Confirm the variance of the adjusted outcome equals the raw variance multiplied by one minus the squared correlation between Y and X, so a correlation of 0.45 removes about 20 percent of the variance and not 80.
Now run that simulation a few hundred times and confirm the adjusted effect estimate is unbiased for the same effect rather than numerically identical to the raw one. Within any single run the two differ, sometimes by a large fraction of the true effect, because the two arms' pre-period covariate means never coincide exactly in a finite sample; they agree in expectation, which is the property that matters and the one to state out loud.

Deliverable: A notebook showing the adjusted estimator with a measurably smaller variance than the raw one, plus a repeated-simulation table showing the two estimators agreeing on average while differing run by run.

Day 4 — Validity threats you can actually test for

Run a sample ratio mismatch check as a chi-square goodness-of-fit test against the intended allocation, and write the three causes you would chase first (assignment logged before exposure, an arm-specific redirect or load failure, bot filtering applied asymmetrically).
Simulate peeking: generate A/A data, test daily at alpha 0.05 across 14 looks, record the inflated false positive rate, then apply an alpha-spending boundary or commit to a fixed horizon and confirm the rate returns to nominal.
Write how you would separate a novelty effect from a durable lift using the treatment effect plotted against days since first exposure, and what shape would change your recommendation.

Deliverable: One table showing the peeking false positive rate before and after correction, plus a written SRM triage list.

Day 5 — When randomization is not available

Write the identifying assumption for difference-in-differences (parallel trends in the absence of treatment), then plot pre-period trends for two candidate control groups and justify rejecting one of them.
Design a switchback test for a change where user-level randomization would leak across participants, choosing a time-block length against the carryover you expect and saying how you would detect carryover in the data.
List what an interrupted time series or a synthetic control buys you and the one thing neither can rule out: an unobserved shock that coincides with the launch.

Deliverable: A one-page memo recommending a single quasi-experimental design and naming its weakest assumption explicitly.

Day 6 — The readout query

Write the assignment-to-exposure join that returns exactly one row per unit per experiment, and handle units appearing in both arms by excluding and counting them rather than silently keeping one.
Compute the per-arm metric, its variance and the relative lift with a confidence interval in SQL, then reproduce the identical numbers in a notebook as a cross-check.
Add a segment breakdown and write the sentence that keeps it from being p-hacking: segments declared in advance, everything else reported as exploratory and corrected for multiplicity.

Deliverable: A single query that outputs the full readout table, matched to a notebook recomputation.

Day 7 — Present it to someone who will not read the appendix

Give a 10-minute readout of a real or simulated experiment in the order decision, number, uncertainty, caveat.
Have your listener ask "can we ship it" in the case where the primary is flat and a guardrail moved, and answer with a recommendation rather than a request for more data.
Rewrite your opening line so the recommendation lands before any methodology.

Deliverable: A one-page readout whose first line is the recommendation.

Ambiguous requests are the test

Most of the questions in this section reduce to one thing: can you be handed a vague request and come back with something useful? Prepare an example where the ask was underspecified, you chose an interpretation, and you said out loud which interpretation you chose. Describing how you narrowed the question matters more than the technique you eventually used.

Share an example of how you used feedback from a peer or mentor to improve a project’s outcome.

Medium · behavioural and stakeholder questions

Approach

Pick a story where you drove the decision, not one where you observed it.
State the situation in two sentences and spend the rest on your reasoning.
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?

Turn an ambiguous onboarding question into a measurable metric

Easy · scoping · metric definition · stakeholder

Two days before a planning review, a director asks whether onboarding is working. You have dim_user (account_created_at_utc, signup_surface, is_internal), fct_event (is_core_action, flow_id, flow_instance_id, event_name, occurred_at_utc, received_at_utc) and fct_session. No further meeting with the director is possible before you start work. Deliver three clarifying questions you would send in writing, the metric you will compute in the meantime with its numerator, denominator, window and exclusions, and one sentence naming the question you are deliberately not answering.

Approach

Recognise what is being probed: whether you convert a goal into a computable predicate without stalling for requirements or guessing in silence. Listing clarifying questions is the generic answer; shipping a defensible default alongside them is the strong one, because the review is in two days and it will happen with or without you.
Infer the decision behind the request. A question about whether onboarding works, arriving before a planning cycle, usually means whether to staff it next quarter. That points at a rate with visible headroom over several cohorts, not at a descriptive dashboard.
Write the three questions so that each one changes the SQL. Which population, all signups or only self-serve from dim_user.signup_surface. What counts as working, reaching a core action or completing the onboarding flow_id. Against what bar, last quarter's cohorts or a stated target.
Propose the default explicitly: seven-day activation on weekly signup cohorts. Numerator, users with is_core_action = TRUE events on at least two distinct UTC dates inside [account_created_at_utc, account_created_at_utc + 7 days). Denominator, the signup cohort with is_internal = FALSE. Publish with an eight-day lag, and state that the two-distinct-days threshold is a frozen choice rather than a discovery.
Name the exclusion in the same breath as the number. The series shows whether users activate; it does not establish that onboarding caused the level, which needs a staged rollout or an experiment.

Follow-up

The director replies that they meant the onboarding flow specifically, not activation. What changes in the query and in the caveats?
Your cohort metric needs an eight-day lag and the review is in two days. What do you present, and how do you label it?
Two of your three questions come back unanswered. Which one do you refuse to proceed without?

Handle a request for numbers supporting a decision already made

Hard · integrity · framing · stakeholder

A senior leader has already decided to sunset a plan tier and asks you for the analysis showing it is the right call. Accounts on that tier carry 6% of MRR at constant FX and have the highest licensed-seat utilisation in the book. The leader's support matters to your next review cycle, and the decision is being presented in four days. Deliver what you produce, what you decline to produce, and the exact sentence you will say in the meeting where the number appears on a slide.

Approach

Recognise what is being probed: whether you can find the legitimate request inside an illegitimate framing instead of either complying or refusing on principle. The generic answer promises to push back; the strong one produces something genuinely useful and states its limits in the room, without ambushing anybody.
Separate the decision from the justification. Sunsetting the tier may be correct for reasons the data does not hold, such as support cost, roadmap surface area or sales motion. What you decline is a one-sided document. What you produce is the case read both ways, which also happens to be more useful to the leader.
Build the symmetric analysis: MRR at risk at constant FX, the share of affected accounts with a plausible migration path given seats_licensed and billing_term, the recovery rate assumed for that migration and where it came from, and the downside case in which high-utilisation accounts treat the sunset as a reason to re-evaluate the vendor entirely.
Surface the inconvenient fact privately and early. The highest seat utilisation in the book is a retention signal, and the leader should hold it before the room does, so they can incorporate it rather than be caught by it.
Agree the meeting sentence in advance with the leader, so that nobody is surprised. Something to the effect that the tier is 6% of MRR and its accounts are the most heavily used in the book, and that the case for sunsetting rests on cost and focus rather than on revenue. That is true, it supports the decision on its real grounds, and it stops the deck claiming the numbers endorse it.
Decide your own line before you need it: what you will not put your name to, and that the route if asked anyway is your own manager rather than a confrontation in the meeting.

Follow-up

The deck circulates with your analysis included and the downside case removed. What do you do, and by when?
What changes if the honest analysis says the sunset is clearly the wrong call?
How do you write the same memo when the leader is your skip-level and the meeting is tomorrow?