Interview guides · AT&T

AT&T Data Scientist Interview Questions & Guide 2026

Causal inference drills for AT&T data scientist prep: diff-in-diff, matching, instruments, and the limits of observed data.

The 60-second brief

At AT&T, data is the backbone of global connectivity. A Data Scientist at AT&T plays a pivotal role in transforming massive streams of telecommunications data into actionable business intelligence and advanced predictive models. From optimizing network performance and predicting equipment failures to reducing customer churn and personalizing marketing campaigns, your work directly impacts millions of customers and shapes the future of connectivity.

Product-sense cases reward reasoning from a mechanism to a testable prediction. Reciting every metric you can name reads as pattern matching; naming the single quantity that would move if your explanation were true reads as thinking.

Focus areas

Analyse at the account grain, cluster errors
Strip CI, retry and synthetic traffic first
Separate contracted seats from actively used seats

Overview

The scale of data at AT&T is virtually unmatched, requiring data professionals to build robust, scalable solutions. Whether you are joining as an experienced professional or entering through the Technology Development Program (TDP), you will be tasked with solving complex problems that require a blend of statistical rigor, machine learning expertise, and strong business acumen.

This role is not just about writing code; it is about driving strategic decisions. You will collaborate closely with product managers, network engineers, and business leaders to translate complex data insights into clear, actionable strategies that keep at the forefront of the telecommunications industry.

The AT&T interview process

AT&T candidates report 4 rounds · ≈ 3-5 weeks. The stages below are what candidates describe, not a published process.

Online Assessment

Before anything else, this round is a reading test. You are given a small schema and a question phrased in business language, and most of the difficulty sits in the gap between them. Who counts as an active user, does a refunded order still count as an order, is that date column an event time or a load time. Weak answers start typing immediately and compute something precise about the wrong population. Strong ones pin the definition in one sentence, name the column that encodes it, then write the query. On a timed assessment with nobody to tell, write the definition in a comment anyway.

What to demonstrate

Whether an ambiguous term becomes a specific column and filter before any computation happens
Whether you read the schema for keys and cardinality rather than only for column names
Whether the result answers the question at the grain it was asked at, per user or per session or per day

How to prepare

Take three metrics you already use and write down the exact filter and exact grain behind each, then practise stating one of them in a single sentence out loud
On a schema you have never seen, spend the first minute writing what one row of each table means and which key it is unique on, then predict which joins can duplicate rows
Rehearse a version where the definition changes halfway through, and edit the query you have instead of starting over

Recruiter Screen

Data Scientist covers at least four different jobs: experimentation, product analytics, causal work on observational data, and applied modelling that ships into a system. A screening call is the cheapest place to find out which of them is being hired for, and doing that diagnosis openly reads as senior rather than fussy. Ask what the last few pieces of work on the team actually were, and roughly how a week splits between querying, modelling and stakeholder time. Then say which parts of that you have done and which you have not. Claiming the whole range is the fastest way to be caught one round later.

What to demonstrate

Whether you can distinguish the flavours of the role and locate your own experience inside one of them honestly
Whether you name what you have not done instead of stretching to cover every line of the posting
Whether your hard constraints (notice period, location, work authorisation, level) surface now rather than at offer stage

How to prepare

Map the last two years of your time into rough percentages across query writing, experiment design, modelling and stakeholder work, so a question about scope has a real answer
Mark every responsibility in the posting as done, adjacent or new, and prepare one sentence for each adjacent item naming the closest thing you have actually built
Decide which logistics are non-negotiable before the call so you can state them in one sentence rather than negotiating live

Technical Evaluations

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

Behavioral Interviews

Behavioural answers from data candidates get audited in a way that answers from other roles do not. When you say a model lifted retention, the next question is the denominator, the window, and how you knew the lift was not seasonal. So attach the measurement to each claim while you tell it: what the metric was before, over what period, and against what comparison. Numbers with no baseline read as rounded-up memory, and one unsupported figure tends to make the rest of the story sound rehearsed.

What to demonstrate

Whether each impact number arrives with a baseline, a window and a comparison, or as a bare percentage
Whether you can name the method that attributed the effect to your work (an experiment, a staged rollout, a seasonal control) or concede the link was correlational
Whether the magnitudes stay internally consistent when the interviewer multiplies them against the scale you described earlier

How to prepare

For each story, write the impact line as metric, value before, value after, window, and how attribution was established. Any line missing two of those five is a follow-up you will answer badly.
Re-derive one headline number from the source table rather than the deck that reported it. Resume numbers drift upward across retellings.
Decide in advance which figures you cannot share, and prepare the ratio or relative change you can give instead, so a confidentiality limit does not read as evasion.

Where candidates lose points

Comparing accounts that received a sales or customer-success touch against those that did not

Assignment of coverage is deliberate and pulls in both directions at once: the largest accounts get a named owner because they are valuable, and the accounts showing distress get one because they are at risk. The comparison therefore mixes a strong positive selection with a strong negative one, and the naive estimate can come out with either sign depending on which assignment rule dominated during the period examined. Nothing about matching on observed size fixes this, because the risk signal that triggered coverage is usually the same signal that predicts the outcome. It needs either an actual randomised or staggered rollout of coverage, or a design built on a capacity constraint or territory boundary that assigns coverage for reasons unrelated to account health.

Reading consumption metrics before the metering lag window has closed

Usage pipelines land late and correct themselves, which is exactly what is_restated and restated_at record. A dashboard queried on day T sees a partially populated tail for the last several days, so the most recent points always slope downward and always look like a regression. Analysts then explain the artefact, and sometimes ship a change to fix it. Establish the empirical settling time by measuring how much a given usage_date's total moves between first_written_at and its final value, exclude that many trailing days from every reportable figure, and never compare a fresh period against a settled one.

Writing SQL without stating NULL and tie-breaking behaviour

Before calling a query finished, say what it does with NULLs, ties and empty groups. NOT IN against a subquery containing a single NULL returns no rows at all, and RANK, DENSE_RANK and ROW_NUMBER differ precisely on ties, so name which one the question requires.

Reading an observational correlation as a causal effect

Name the confounder you are most worried about and the design that would remove it: an experiment, a difference-in-differences with a checked pre-period trend, an instrument, or a regression discontinuity. When none is available, state which direction the bias likely runs and bound the claim accordingly.

Questions & practice

Explain the mathematical theory and decision boundaries behind a machine learning algorithm you have used in a past project.

Medium · machine learning and modelling

Approach

Say how the offline result would be validated online before it is trusted.
Check what information would not exist at prediction time, and exclude it.
Set a baseline first, so any model has something honest to beat.

Follow-up

What would you monitor after launch to know the model is still valid?
How would you choose the decision threshold, and who owns that choice?

Describe a scenario where you had to choose between a simple interpretable model and a complex black-box model.

Medium · machine learning and modelling

Approach

Check what information would not exist at prediction time, and exclude it.
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

What would you monitor after launch to know the model is still valid?
How would you choose the decision threshold, and who owns that choice?

Solve a basic algorithmic coding problem equivalent to an easy-level data structures challenge.

Medium · machine learning and modelling

Approach

Pick an evaluation metric that matches the cost of each error type, not a default.
Check what information would not exist at prediction time, and exclude it.
Frame the prediction: the label, the moment of prediction, and the action it triggers.

Follow-up

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

Bootstrap a confidence interval for net revenue retention

Medium · bootstrap · retention · resampling

You have one row per account with arr_start_cents (ARR twelve months ago) and arr_end_cents (ARR today, zero if churned), covering the fixed cohort of accounts that had ARR twelve months ago. Net revenue retention is sum(arr_end_cents) / sum(arr_start_cents). Write a nonparametric bootstrap from scratch, without scipy.stats.bootstrap: resample accounts with replacement, recompute the ratio of sums on each resample, and return the point estimate with a 95% percentile interval from 10,000 resamples. Also report the interval you would get from the mean of per-account ratios, and explain the difference.

Approach

Resample the account, because the account is the unit the estimand is defined over. One bootstrap draw is a vector of account indices and both numerator and denominator are recomputed from that same draw; resampling the two sides independently destroys the within-account correlation that makes a ratio estimator stable.
Vectorise the draws: idx = rng.integers(0, n, size=(B, n)), then end[idx].sum(axis=1) / start[idx].sum(axis=1). A 10,000 by n index matrix is usually far cheaper than a Python loop; if the matrix is too large for memory, chunk over B rather than reverting to a loop.
Take the interval from np.quantile(ratios, [0.025, 0.975]). The percentile interval differs from estimate +/- 1.96 * bootstrap SE whenever the resample distribution is skewed, which it will be here, and the skew is the thing you want represented.
Compute the mean-of-ratios version on the same resamples, and state the exact relationship rather than guessing which of the two is larger. With r_i = arr_end_i / arr_start_i, the ratio of sums is the arr_start-weighted mean of exactly those r_i, so sum(end)/sum(start) - mean(r) = Cov(arr_start, r) / mean(arr_start) using the population covariance. The gap is positive when larger accounts retain and expand better than smaller ones, and negative when they do not; a cohort whose small accounts churn at a higher rate has positive covariance, which puts the mean of per-account ratios BELOW the ratio of sums. Requires arr_start_i > 0 for every account, which the fixed-cohort definition guarantees; r_i is floored at 0 and unbounded above, so a handful of 4x expansions among small accounts can flip the sign. Compute the covariance and report it instead of asserting a direction.
Report the interval width beside the concentration of the cohort. If the largest account is 12% of starting ARR, a narrow interval is evidence that the resampling unit is wrong rather than evidence that the estimate is precise.

Worked solution (30 min)

start = df.arr_start_cents.to_numpy(float); end = df.arr_end_cents.to_numpy(float); n = len(start); point = end.sum() / start.sum()
rng = np.random.default_rng(7); idx = rng.integers(0, n, size=(10_000, n)); ratios = end[idx].sum(1) / start[idx].sum(1)
lo, hi = np.quantile(ratios, [0.025, 0.975]); return point, lo, hi
per_acct = end / start; mean_point = per_acct.mean(); mean_boot = per_acct[idx].mean(1); compare np.quantile(mean_boot, [0.025, 0.975]) against (lo, hi), and report np.cov(start, per_acct, ddof=0)[0,1] / start.mean() as the quantity that accounts for the gap between the two centres.

Expected result: A point estimate equal to sum(end)/sum(start) on the unresampled data — the bootstrap mean is close but must not be substituted for it — and a percentile interval that is asymmetric around the point when expansion is skewed. The mean-of-ratios interval is centred on a different estimand, and which side it falls on is decided by Cov(arr_start, r): it sits below the ratio of sums in the common case where small accounts churn at a higher rate than large ones, and above it when the large accounts are the ones shrinking. Its width is governed by the spread of per-account growth, while the ratio-of-sums width is governed by how concentrated starting ARR is, so neither interval is reliably the wider one.

Verify it yourself

point equals sum(end)/sum(start) computed directly on the original arrays, and point - per_acct.mean() equals np.cov(start, per_acct, ddof=0)[0,1] / start.mean() to floating-point tolerance. That is an algebraic identity, not an empirical claim: a mismatch means the per-account ratios or the weights are wrong, and the sign of it tells you which of the two numbers is larger before you look.
np.median(ratios) sits within a small fraction of a point of the point estimate; a large gap indicates a wrong resampling unit or a transposed index matrix.
Raising B from 10,000 to 20,000 moves the endpoints by far less than the interval width.
Setting every arr_end equal to its arr_start gives point == 1.0, per_acct.mean() == 1.0, a zero covariance term and a zero-width interval.

Follow-up

The cohort has 800 accounts and the largest is 12% of starting ARR. How much do you trust a percentile interval here?
How would you extend this to an interval on the year-over-year change in NRR?
Two accounts merged mid-window and one contract was co-termed into the other. How do you keep the cohort fixed?

Explain how you would merge two large datasets in Python without running out of memory.

Medium · sql and data manipulation

Approach

State the window function and its partition and ordering out loud before writing it.
Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.
Say which table is the grain you start from, and join outward from it.

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?

Write a SQL query using GROUP BY and JOIN to aggregate customer usage metrics.

Medium · sql and data manipulation

Approach

Check whether any join is one-to-many before aggregating, or the sums inflate.
Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.
Say which table is the grain you start from, and join outward from it.

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?

Sessionise interactive API traffic with a thirty-minute inactivity gap

Medium · window functions · sessionisation · gaps and islands

fct_api_request carries account_id, user_id (null for service accounts), request_at, traffic_class and http_status. Using only traffic_class = 'interactive' rows with a non-null user_id, group each seat's requests into sessions on a 30-minute inactivity threshold: a request more than 30 minutes after the previous request from the same (account_id, user_id) opens a new session. For one ISO week return, per account, the session count, the median session duration in minutes and the median requests per session. A single-request session has a duration of zero.

Approach

Filter first: traffic_class = 'interactive' and user_id IS NOT NULL. Machine traffic has no sessions in any useful sense, and leaving CI or batch rows in produces sessions that are really cron schedules.
Get the previous timestamp with LAG(request_at) OVER (PARTITION BY account_id, user_id ORDER BY request_at). Partitioning by user_id alone stitches one person's work across two different accounts into one fabricated session, because a human holds memberships in several accounts.
Flag a boundary where the lag is NULL or request_at - lag > interval '30 minutes'. Decide and state whether exactly 30 minutes continues the session; strictly greater is the conventional choice and needs to be written down either way.
Assign session ids with SUM(boundary::int) OVER (PARTITION BY account_id, user_id ORDER BY request_at ROWS UNBOUNDED PRECEDING), the standard running-count construction for islands.
Roll up to sessions with min(request_at), max(request_at) and count(*), then to accounts with percentile_cont(0.5) WITHIN GROUP (ORDER BY ...). Use medians, not means: session length is strongly right-skewed and one long-running client dominates the average.

Worked solution (30 min)

Build the filtered week: interactive rows with user_id IS NOT NULL inside [week_start, week_start + 7 days).
Add prev_at via LAG(request_at) OVER (PARTITION BY account_id, user_id ORDER BY request_at) and derive is_new_session = (prev_at IS NULL OR request_at - prev_at > interval '30 minutes').
Add session_seq = SUM(is_new_session::int) OVER (PARTITION BY account_id, user_id ORDER BY request_at ROWS UNBOUNDED PRECEDING).
Aggregate to sessions by (account_id, user_id, session_seq) taking min, max and count, with duration_minutes = EXTRACT(EPOCH FROM max - min) / 60.
Aggregate to accounts with count(*) AS sessions, percentile_cont(0.5) WITHIN GROUP (ORDER BY duration_minutes) and percentile_cont(0.5) WITHIN GROUP (ORDER BY request_count).

Expected result: One row per account with session_count, median_session_minutes and median_requests_per_session, computed over interactive human traffic only, where a lone request yields a session of duration zero and request count one.

Verify it yourself

Summing request_count across all sessions equals the filtered row count for the week exactly.
Every session duration is non-negative, and no two consecutive requests inside one session are more than 30 minutes apart when you re-derive the LAG within sessions.
Recomputing with the partition on user_id alone yields strictly fewer sessions and longer medians for any account sharing users with another account, which confirms the partition key matters.

Follow-up

Sessions belonging to colleagues in one account are correlated. What does that do to a t-test on session length across an experiment arm?
A long-poll or streaming endpoint keeps a connection open for hours. How do you stop it reading as one twelve-hour session?

Design a seat-utilisation metric before enforcing licensed seat limits

Medium · seat utilisation · denominator design · service accounts · renewal timing

Seat limits are currently unenforced. Leadership wants a metric to decide whether to enforce them at renewal. You have dim_user_membership (user_id, account_id, role, seat_type, is_service_account, activated_at, deactivated_at, last_seen_at) and fct_subscription_period (account_id, contracted_seats, pricing_model, term_end_date, arr_cents, is_current). Define licensed seat utilisation at the account grain, state every exclusion in both the numerator and the denominator, and say what a value above 1.0 means. Deliver the metric plus the guardrail that protects against enforcement destroying more ARR than it recovers.

Approach

Numerator: distinct user_id in dim_user_membership with seat_type = 'licensed_paid', is_service_account = false, deactivated_at null and last_seen_at inside a trailing 28 days. Service accounts hold a membership and burn API keys but consume no human attention, so counting them inflates utilisation hardest on the most deeply integrated accounts, which is exactly backwards.
Denominator: contracted_seats from the fct_subscription_period version current at the evaluation date. Exclude accounts with pricing_model = 'pure_consumption' entirely, because contracted_seats is null there and the ratio is undefined rather than zero. A null coerced to zero produces an infinite ratio that will lead the report.
Interpret both tails and say what each means commercially. Above 1.0 means the entitlement is not enforced and humans are sharing seats, which is recoverable revenue. Persistently below roughly 0.5 is the most reliable available signal of a seat reduction at renewal, because buyers audit seat counts at renewal and at no other moment.
Guardrail: ARR at risk from enforcement, the sum of arr_cents over accounts with utilisation above 1.0 whose term_end_date falls inside the enforcement window. Enforcing on an account that responds by cutting seats at renewal converts a recoverable overage into a permanent downgrade, and the recovered revenue can be smaller than the loss.
Report the metric as a distribution with the two tails counted separately and the renewal calendar attached, because no account can act on enforcement until its own term ends. A fleet average of seat utilisation is a number nobody can act on.

Worked solution (20 min)

Build the current contract per account from fct_subscription_period where is_current = true, keeping contracted_seats, pricing_model, term_end_date and arr_cents, and drop pure_consumption rows.
Count active humans per account from dim_user_membership under the four filters, using distinct user_id per account_id, which is safe because the table is one row per membership and not one row per human.
Compute the ratio and bucket accounts into below 0.5, 0.5 to 1.0, and above 1.0, joining arr_cents and term_end_date.
Report ARR in each bucket and the share of that ARR renewing in the next two quarters.

Expected result: Three utilisation buckets with ARR and renewal timing attached, and a metric definition whose numerator excludes service accounts, guests, free collaborators and deactivated memberships, and whose denominator is contracted_seats from the contract version current at the evaluation date, with pure-consumption accounts excluded rather than counted as zero.

Verify it yourself

No account with pricing_model = 'pure_consumption' may appear, with a null or an infinite ratio.
Bucket ARR must sum to the total ARR over included accounts.
Summing per-account active-user counts across the fleet must be greater than or equal to the fleet-wide count of distinct user_id, because one person can hold memberships in several accounts; exact equality means multi-account users were deduplicated at the wrong grain.

Follow-up

last_seen_at is null for service accounts by definition. If the is_service_account flag is unreliable, what happens to your numerator, and how would you audit the flag?
Why a 28-day activity window rather than 7 or 90, and what evidence would change your answer?

Separate novelty from durable lift in a console redesign

Medium · novelty-effect · time-heterogeneity · holdout

A redesigned request console was tested for four weeks, randomised at the account level. The lift in weekly successful interactive requests per account was +9.2%, +5.1%, +1.8% and +0.4% in experiment weeks one through four, and the week-four 95% interval spans plus or minus 3.1 percentage points. Accounts entered on a rolling basis as they next signed in, so week one is a different calendar week for different accounts. Say what the decaying series can and cannot establish, specify the readout that would settle it, and state what you would ship on Monday.

Approach

Name the distinct explanations the same curve is consistent with, because the shipping decision differs across them: a novelty effect decaying to zero; a novelty effect decaying to a small positive plateau the test has no power to see; and a genuine effect whose measured size shrinks as control accounts learn about the change from colleagues or release notes.
Fix the time axis before interpreting anything. With rolling entry, calendar week and weeks-since-first-exposure are different variables, and pooling them mixes the decay curve with a change in cohort composition, since the accounts that signed in on day one are systematically the most engaged. Recut on weeks-since-first-exposure and verify that plan_tier and pre-period usage mix are stable across entry cohorts.
Read the interval width honestly. Plus or minus 3.1 percentage points at week four cannot separate a durable +1% from zero. The defensible statement is that no durable effect larger than roughly 3.5% was detected, not that there is no durable effect, and those two sentences lead to different decisions.
Separate novelty from primacy using account tenure. Accounts created after launch have no prior console to be surprised by, so they cannot show novelty; if the redesign is genuinely better their curve should be flat or rising, while established accounts show the spike and decay. If both cohorts decay to zero, the effect really was novelty.
Specify the settling readout rather than arguing about the four weeks you have. Hold back 5% of accounts as a long-run holdout for 90 days, and compare whole ISO weeks only: usage in this domain follows a hard five-to-two weekday cycle, and a window containing four business days instead of five moves this metric by several percent with no product change.
Give the Monday answer. Ship if guardrails are clean and maintenance cost is low, because a decayed-to-zero effect with no harm is a neutral trade; but book none of the +9.2% in any forecast, and do not run a follow-up test on the same accounts inside the novelty window, because their baseline has not returned to steady state.

Follow-up

What sample or horizon would you need to rule out a durable +1.5% at 80% power, given the week-four interval you have?
How would you distinguish a novelty effect from a control arm that gradually learned about the change?
The metric that decides renewal is up to eleven months away on annual contracts. What proxy do you use in the meantime, and how do you validate it once renewals land?

Consumption per account always dips in the last four days

Easy · pipeline lag · restatement · reporting windows

Your consumption dashboard reads billable units per paying account from fct_usage_daily (account_id, sku_code, usage_date, billable_quantity, net_amount_cents, is_restated, first_written_at, restated_at, updated_at). Every refresh shows the final three to five days declining, and a review is scheduled on that shape. Establish empirically how long a usage_date takes to settle, separately per sku_code, and specify the trailing exclusion window every reported figure should use. Derive the number from the data; do not adopt a convention.

Approach

Treat settlement as a measurable curve rather than a belief: for each historical usage_date, compare the total captured as of age d days against that date's final settled total.
Do it per sku_code, because metering paths differ. A per-request SKU lands within hours, while a storage-month SKU is produced by a daily sweep and lands later, so one global lag number is wrong for at least one of them.
Pick the age at which a stated percentile of dates reaches a stated completeness threshold, for example the tenth percentile date reaching 99.5 percent of final, and take the slowest SKU's age as the dashboard's exclusion window.
Separate late arrival from restatement. Late rows raise totals and are fixed by waiting; restatements can move either way and are not, so report the share of the gap from each.
Encode the exclusion inside the query rather than in the chart, so anyone reusing the SQL inherits the rule, and never compare a fresh partial period against a settled one.

Follow-up

Month-end invoicing needs a number before settlement completes. How would you publish an early estimate with an honest uncertainty band attached?
What monitor would tell you the settling time has changed, without anyone remembering to re-run this analysis?

Suggested practice checklist

Four days spend equal time on query work, statistics, modelling and product judgement at deliberately shallow depth, which produces a scored map of where you actually stand. The last three days spend everything on the two areas the role weights most, and close by re-running day one to measure movement.

Day 1 — Breadth pass: query fluency

Solve six prompts spanning aggregation, joins, window functions and date arithmetic in 60 minutes total, stopping at 10 minutes each whether or not it works, and mark every prompt as solved, solved slowly, or stuck.
For each unsolved prompt write the single blocking sentence (I lost the grain, I did not know the frame clause, I could not express the date boundary) instead of reading the solution.
Translate one pandas transformation you know well into SQL and one SQL query into pandas, checking that both return the same row count and the same totals.

Deliverable: A scored six-row table, one line per prompt, saved for the day-seven re-run.

Day 2 — Breadth pass: statistics and inference

Answer ten short questions in writing with nothing open: what a p-value is conditional on, what a 95 percent interval covers across repeated samples, when a paired test is the right one, what the bootstrap estimates, why multiple comparisons inflate false positives, how controlling the family-wise error rate differs from controlling the false discovery rate, what power depends on, what a missed real effect costs a product, the three situations where the central limit theorem does not rescue you (small n, very heavy tails, dependent observations), and what a standard error is the standard deviation of.
Grade yourself against a reference and count only the answers that were exactly right, not the ones that were nearly right.
Rewrite the two weakest answers the following morning from memory in full sentences.

Deliverable: Ten graded answers with an honest count of exact hits.

Day 3 — Breadth pass: modelling

Take one tabular dataset end to end in 90 minutes: a leakage-safe split, a baseline that is not a model (majority class or historical mean), one regularized linear model, one gradient-boosted tree, and a single evaluation metric chosen before you look at any result.
Write why that metric fits the cost structure: precision at a fixed recall for alerting, calibration for anything feeding a price or a threshold, ranking metrics for retrieval, and note that area under the ROC curve is insensitive to class balance in a way that can flatter a rare-positive problem.
Name the leak you were most likely to introduce (an encoding fit on all rows before splitting, or a feature computed after the label's timestamp) and write the check that would have caught it.

Deliverable: A notebook whose first cell states the metric and the baseline, plus two lines on what beat what and by how much.

Day 4 — Breadth pass: product judgement

Answer three case prompts aloud at 15 minutes each, timing how long passes before you state a success metric.
For one case write the first segmentation you would run and the row counts you expect per segment, so that a tiny segment cannot quietly drive the conclusion.
Take a metric definition you did not write, from a public dashboard, a textbook, or documentation you already have open, and list every place two analysts implementing it would diverge: which rows the denominator admits, whether the unit is an account or a person, what the time window is anchored to, and what happens to data that arrives late. Then write the one question that would close the largest of those gaps.

Deliverable: Three recorded case answers plus an ambiguity list for a metric someone else defined, ending in the single question you would ask about it.

Day 5 — Depth, first area

Rank the four areas by how many bullet points in the role description each one covers, pick the top one, and spend the entire day inside it.
Work the six hardest problems you can find in that area and for each write the generalizable move you should have reached for first, rather than the answer.
Re-solve the two you failed the same evening with notes closed.

Deliverable: Six generalizable moves written as instructions to yourself, not as solutions.

Day 6 — Depth, second area, and the seam between them

Repeat the depth protocol on the second-ranked area with the same six-problem structure.
Construct one problem that requires both areas at once, for example a metric redefinition whose effect you must validate with a test whose readout you then have to query.
Solve your own combined problem end to end and note where the handoff between the two areas cost you time.

Deliverable: One combined problem, solved end to end, with the handoff failure written down.

Day 7 — Integration and re-measurement

Re-run the six prompts from day one under the same clock and compare both correctness and time.
Run a 60-minute mixed mock that moves between areas without warning, since switching cost is what breadth passes do not train.
Write the two areas you would still fail on, and the sentence you will use in the interview when you hit one of them.

Deliverable: A before-and-after score table plus a written plan for the two remaining gaps.

The project you argued against

Saying no well is a senior skill and it is rarely rehearsed. Think of a time you told someone their analysis was not worth doing, or that the experiment could not answer their question at the sample size available. Explain what you offered instead. Refusal without an alternative reads as obstruction rather than judgement.

Describe a situation where you faced a significant technical challenge during a project. How did you identify the root cause, and what steps did you take to fix it?

Medium · behavioural and stakeholder questions

Approach

State the situation in two sentences and spend the rest on your reasoning.
Pick a story where you drove the decision, not one where you observed it.
Name the disagreement or constraint, and how you resolved it with evidence.

Follow-up

What would you do differently if you ran that project again?
What did you decide not to do, and why?

What is your greatest professional or academic accomplishment in the field of data science?

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.
Name the disagreement or constraint, and how you resolved it with evidence.

Follow-up

How did you know the outcome was caused by your change?
What did you decide not to do, and why?

Walk through an analysis you later discovered was wrong

Easy · data quality · error ownership · metering

Six weeks ago you reported that consumption fell 9 percent in the last week of the month, and a team spent a sprint investigating the cause. The fall was an artefact: rows in fct_usage_daily land late and are restated in place, and you queried before the tail had settled. Describe how you found the error, what you told the people who acted on it, and the control you put in place so this class of mistake cannot reach a dashboard again. Be specific about how the settling window was measured.

Approach

The interviewer is probing whether you self-report errors before someone else finds them, and whether your fix is structural rather than a promise to be more careful. Say plainly that the number was wrong and that a sprint was spent on it, before describing any diagnosis.
Establish the artefact quantitatively instead of asserting that data lands late. For each usage_date, compare the total as of first_written_at against the settled total and read the settling time off that curve, for example 97 percent of final by day three and 99.5 percent by day five.
Correct the record the same day, in the channel the original number went out in, to the same audience. The cost of the wasted sprint belongs in the correction, not in a footnote.
Make the fix structural: exclude a trailing lag window from every reportable figure, and make the reporting view return no rows inside that window rather than returning partial ones. A dashboard that shades unsettled days still gets read as a decline.
State what generalises. Any fact table restated in place has this failure mode, so the guard belongs at the source rather than on the one dashboard that embarrassed you. A strong answer ends with the class of error closed; a generic one ends with a lesson learned.

Follow-up

How did you choose the completeness threshold behind the lag window, and what would make you recalibrate it?
What did you say to the team that lost the sprint, and what did they say back?
Is there a legitimate case for showing the unsettled tail at all, and to whom?