Interview guides · Rhombus Power

Rhombus Power Data Scientist Interview Questions & Guide 2026

SQL practice for Rhombus Power data scientist candidates: window functions, funnel and retention queries, joins that quietly drop rows.

The 60-second brief

A Data Scientist at Rhombus Power sits at the critical intersection of artificial intelligence, large-scale data engineering, and national security. The primary mission of this role is to design, develop, and launch production-ready data pipelines and quantitative models that deliver real-time, AI-powered predictive decision support for defense and national security organizations. This is not a purely theoretical research role; it is an engineering-heavy data science position where your models will directly influence strategic decision-making in high-stakes environments.

In modelling rounds the live question is usually why this model class for this problem, and how you would know six weeks after launch that it is still working. Deriving gradients by hand is rarely what is being probed.

Focus areas

Power experiments for heavy-tailed account revenue
Separate contracted seats from actively used seats
Analyse at the account grain, cluster errors

Overview

In this role, you will work on complex, multi-faceted datasets to build solutions for critical partners, such as the Air Education & Training Command and other defense sectors. The systems you build must handle massive-scale data processing with absolute reliability, often integrating into high-side (classified) secure environments. You will be responsible for the entire lifecycle of data—from discovering and curating unstructured datasets to deploying and validating robust predictive models that run on high-performance computing (HPC) or cloud architectures.

The impact of your work at Rhombus Power is immediate and profound. By transforming raw, disparate data into actionable intelligence, you enable decision-makers to anticipate threats, optimize resources, and execute missions with unprecedented speed and precision. This requires a unique blend of mathematical rigor, software engineering discipline, and a deep commitment to national security missions.

The Rhombus Power interview process

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

Technical Screen

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

Take-Home Challenge

Before any modelling, the dataset is itself the first test. Take-home data usually carries something broken: rows duplicated at an unexpected grain, a join that silently drops part of the population, timestamps stored in more than one timezone, or missingness correlated with the outcome. An hour spent profiling row counts, key uniqueness and date ranges is not overhead, because it decides whether every number after it is real. What separates submissions is whether you report the defects you found and adapt the analysis to them, rather than modelling over them quietly and hoping the aggregate absorbs it.

What to demonstrate

Whether you established the grain of each table and checked row counts after every join, and said so in the writeup
Whether data defects you found are surfaced with their effect on the conclusion, instead of being dropped without comment
Whether filters and exclusions are reproducible from the submitted code, with the size of the excluded population quantified

How to prepare

Write a short profiling script you can point at any unfamiliar table: row count, distinct key count, null rate per column, and the min and max of every date field, then run it before anything else
Write the funnel or the join chain as one query and check the row count at each grain, so a silent fan-out shows up as a number rather than as a wrong answer later
On a past dataset, list every exclusion you applied and how many rows each one removed, then draft the single sentence about it you would put in a report

Solution Presentation

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.

Behavioral and Technical 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

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.

Reporting a mean over accounts when account revenue is heavy-tailed

When a small number of accounts hold most of the revenue, the sample mean is dominated by whichever of them happens to be in the sample, and the sample variance keeps growing as more data arrives instead of stabilising. In that regime the usual central-limit-based confidence interval understates uncertainty, and a single renewal or a single large account's batch job can flip the sign of a measured effect. The fixes are to pre-register a winsorisation or capping rule before looking at the outcome, to report account counts crossing a threshold alongside the revenue figure, or to define the estimand on a bounded transform. Choosing the cap after seeing the result is a separate and worse problem, because the cap then encodes the answer.

Reporting a mean for a heavy-tailed metric without saying what it hides

For spend, session length or items per order, a small fraction of units carries most of the total, so the mean has a wide standard error and one account can move it. Fix the handling before you see the result: cap or winsorise at a pre-declared percentile, and report the median or the share above a threshold next to the mean. Capping changes the estimand, so say which question the capped number answers, and check how much of any difference comes from the top 0.1 percent of units.

Explaining an aggregate move without decomposing the mix shift

Split the change in the aggregate into within-segment movement and movement in segment weights before you explain it. Every segment's rate can fall while the overall rate rises, purely because volume shifted toward segments that already had higher rates.

Questions & practice

Walk me through the technical architecture of the most complex machine learning project on your resume. Why did you make those specific design choices?

Medium · machine learning and modelling

Approach

Frame the prediction: the label, the moment of prediction, and the action it triggers.
Check what information would not exist at prediction time, and exclude it.
Say how the offline result would be validated online before it is trusted.

Follow-up

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

Why would you choose a specific activation function (like ReLU or GeLU) over a default activation function in a deep neural network project?

Medium · machine learning and modelling

Approach

Check what information would not exist at prediction time, and exclude it.
Say how the offline result would be validated online before it is trusted.
Pick an evaluation metric that matches the cost of each error type, not a default.

Follow-up

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

Quantify billable volume created by retries after server errors

Hard · telemetry-hygiene · pandas · unit-economics

From fct_api_request (request_id, account_id, endpoint, idempotency_key, is_retry, http_status, request_at, billable_units), measure how much billable volume in a 28-day window is retry traffic that followed a 5xx. Group requests into attempt chains by (account_id, endpoint, idempotency_key) ordered by request_at; a request is error-driven if any earlier attempt in its chain returned 5xx. Requests with a null idempotency_key cannot be chained, so report them as their own class rather than assuming each is unique. Return billable_units split into first-attempt, error-driven retry, other retry and unchainable, per account.

Approach

Split the population before measuring anything. A null idempotency_key is not a chain of one, it is an unknown; report its share of billable_units first, because if it is 40% of volume then the headline estimate is a lower bound and the deliverable has to say so.
Within chainable rows, sort by (account_id, endpoint, idempotency_key, request_at) and derive 'any earlier attempt failed' with arithmetic rather than a per-group lambda: with is5 = (http_status >= 500), the per-chain cumsum minus the row's own value is positive exactly when an earlier attempt in that chain returned 5xx. A groupby-apply gives the same answer and is unusable at five million rows.
Do not take is_retry as the definition. It is set by the client whenever an idempotency_key is resent, which covers retries after client-side timeouts and after 4xx as well; compute the flag yourself and then cross-tabulate it against is_retry, because the disagreement is a finding in its own right.
Aggregate billable_units by (account_id, class) and assert the classes sum to each account's total. The spine sets billable_units to zero on 5xx responses, so the failed attempt contributes nothing and the whole inflation sits in the successful retry that follows it.
Report the per-account share and look at its distribution, not the fleet total. One account in a retry storm dominates any blended figure, which is the same failure that makes a fleet-wide error rate useless.

Worked solution (40 min)

unchainable = df.idempotency_key.isna(); report df.loc[unchainable].groupby('account_id').billable_units.sum() before proceeding.
keys = ['account_id','endpoint','idempotency_key']; c = df[~unchainable].sort_values(keys + ['request_at'], kind='mergesort'); c['is5'] = (c.http_status >= 500).astype(int)
c['attempt_no'] = c.groupby(keys, sort=False).cumcount(); c['prior_5xx'] = (c.groupby(keys, sort=False).is5.cumsum() - c.is5) > 0
c['cls'] = np.where(c.attempt_no == 0, 'first_attempt', np.where(c.prior_5xx, 'error_driven_retry', 'other_retry')); out = pd.concat([c, df[unchainable].assign(cls='unchainable')]).groupby(['account_id','cls']).billable_units.sum().unstack(fill_value=0)

Expected result: A per-account table of four columns summing exactly to that account's total billable_units in the window. error_driven_retry is zero for any account with no 5xx; the fleet-level error-driven share is typically low single digits while a small number of accounts sit far above it.

Verify it yourself

out.sum(axis=1) equals df.groupby('account_id').billable_units.sum() for every account.
Every row classed error_driven_retry has attempt_no >= 1.
An account with no http_status >= 500 in the window has error_driven_retry == 0.
Cross-tabulating cls against is_retry: no first_attempt row should carry is_retry True, and any that do mark chains truncated by the window boundary.

Follow-up

An account's error-driven share is 22%. Is that the platform's fault or the client's, and what do you look at next?
How would you define a consumption-based north-star metric that an outage cannot inflate?
Chains straddle the 28-day boundary. How large is that bias and in which direction?

Explain the key differences between memory management in Pandas and PyArrow when handling large datasets.

Medium · sql and data manipulation

Approach

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.
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?
What breaks if events arrive late or out of order?

How would you optimize a slow-running SQL query that joins multiple large-scale tables containing millions of rows?

Medium · sql and data manipulation

Approach

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

What breaks if events arrive late or out of order?
How would you verify this result without re-running the same query?

Explain how Python manages memory under the hood, specifically focusing on garbage collection and reference counting.

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 would you verify this result without re-running the same query?
What breaks if events arrive late or out of order?

How do you sort a list of tuples in Python based on the second element? Can you write this solution both with and without using a lambda function?

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?
What breaks if events arrive late or out of order?

Find paid seats that never called the API

Easy · anti-join · null semantics · seat utilisation

dim_user_membership carries user_id, account_id, seat_type, is_service_account, deactivated_at and last_seen_at. fct_api_request carries user_id, which is null whenever the caller is a service account or an unattended key, plus account_id, request_at and http_status. For a single account, list every licensed_paid seat with is_service_account = false and deactivated_at null that issued no request at all in the trailing 90 days, returning user_id and last_seen_at. A first draft writes NOT IN against a subquery over fct_api_request.user_id. State exactly what that draft returns and why, then write the correct query.

Approach

Read the column comment first: user_id is nullable on the fact, so the subquery almost certainly contains at least one NULL for any account that runs a service account or an unattended key.
Work the three-valued logic out loud. x NOT IN (a, NULL) expands to NOT (x = a OR x = NULL); the second disjunct is UNKNOWN, so for any x not equal to a the whole predicate is UNKNOWN and the row is filtered out. The draft returns zero rows, which reads as full seat utilisation.
Rewrite as NOT EXISTS with the 90-day and status predicates inside the correlated subquery. Putting them in the outer WHERE instead turns the anti-join into a different question and silently changes the answer.
Correlate on both user_id and account_id. A membership is (user_id, account_id) and one person can hold memberships in several accounts, so correlating on user_id alone marks a seat as active because that human was busy somewhere else.
Filter the seat side to seat_type = 'licensed_paid', is_service_account = false and deactivated_at IS NULL, and say what the resulting count means next to contracted_seats on the current subscription row.

Worked solution (15 min)

Confirm the hazard with one query: SELECT count(*) FROM fct_api_request WHERE user_id IS NULL AND account_id = :account_id. Any non-zero result proves the draft returns nothing.
Write the seat side: SELECT user_id, last_seen_at FROM dim_user_membership WHERE account_id = :account_id AND seat_type = 'licensed_paid' AND NOT is_service_account AND deactivated_at IS NULL.
Attach AND NOT EXISTS (SELECT 1 FROM fct_api_request r WHERE r.user_id = m.user_id AND r.account_id = m.account_id AND r.request_at >= now() - interval '90 days').
Compare the row count against the same query written as a LEFT JOIN with a WHERE r.user_id IS NULL; the two must agree exactly.
Divide the active seat count by contracted_seats from the account's current fct_subscription_period row and state the utilisation figure.

Expected result: A list of user_id and last_seen_at for licensed, non-service, active memberships on that account with no request in 90 days. The NOT IN draft returns zero rows whenever the account has any service-account or unattended-key traffic, and returns the right answer only by accident on accounts that have none.

Verify it yourself

The NOT EXISTS result and the LEFT JOIN ... IS NULL result return identical user_id sets.
Idle seats plus seats with at least one request in 90 days equals the total licensed non-service active membership count for the account.
Forcing a NULL into the subquery (union a SELECT NULL) leaves NOT EXISTS unchanged and empties the NOT IN version, which demonstrates the failure on demand.

Follow-up

Adding AND user_id IS NOT NULL to the subquery also fixes NOT IN. Why is NOT EXISTS still the form you would leave in the repository?
last_seen_at looks like a shortcut for the whole query. What does it actually record, and where does it disagree with API activity?
This is a seat-reduction risk list. What threshold would you attach before handing it to an account team, and what happens to seats below it?

How do you ensure your models remain performant and reliable when deployed in highly restricted, secure environments?

Medium

Approach

Say what you would check first and why it is the highest-information step.
Work from the decision backwards to the evidence you would need.
State your assumptions explicitly before working the problem.

Follow-up

What assumption would you test first?
How would you know your answer was wrong?

Size an account-randomised test of a consumption feature

Easy · power · mde · heavy-tailed-outcomes

You have 2,800 non-internal paying accounts. The outcome is the monthly sum of billable_quantity for sku_code = 'compute_hours' from fct_usage_daily, with a mean of 640 and a standard deviation of 2,100 across accounts. Product wants to detect a 5% lift from a scheduling feature, randomised 50/50 at the account level, at 80% power and two-sided alpha 0.05. Give the minimum detectable effect the available sample actually supports, say whether the 5% target is reachable, and state what you would change in the design or the metric before agreeing to run it.

Approach

Write the two-sample formula and plug it before discussing anything else: MDE = (z_0.975 + z_0.80) * sd * sqrt(1/n1 + 1/n2) = 2.80 * 2100 * sqrt(2/1400) = 222 compute-hours, which is 34.7% of the 640 mean. The driver is the coefficient of variation of 3.3, not the account count, so quoting a sample size without it is not a power calculation.
Invert for the stated target. A 5% lift is 32 compute-hours, needing N = 4 * (2.80 * 2100 / 32)^2 = about 135,000 accounts, roughly 48 times the eligible population. Say the test is not powerable on this metric, and do not offer to run it longer: the account population is fixed, and extra weeks add repeated observations on the same clusters rather than new clusters.
Attack the variance instead of the sample. Pre-register winsorisation at the 99th percentile of the pre-period distribution, or move the estimand to a bounded one such as the share of accounts exceeding a usage floor. State explicitly that this changes the estimand rather than cleaning the data, because a cap chosen after seeing the result encodes the answer.
Add CUPED on the account's pre-period compute_hours. At a pre/post correlation of 0.80 the residual variance multiplier is 1 - 0.80^2 = 0.36, the standard error falls to 0.60 of its unadjusted value, and the MDE drops from 222 to 133 compute-hours (20.8% relative). That is still four times the requested 5%.
Fix the randomisation unit and the balance in the same breath: randomise account_id because colleagues share workspaces, and stratify assignment on pre-period usage decile so that a handful of very large accounts cannot land disproportionately in one arm and dominate the difference in means.
Close with an operating decision rather than a number. Run only if the team will act on a roughly 20% MDE after CUPED and winsorisation; otherwise re-specify the primary outcome as a bounded indicator and treat compute-hours as a secondary reported with intervals only.

Worked solution (25 min)

Compute the coefficient of variation, 2100 / 640 = 3.28, and check the top-decile share of total compute_hours before treating the mean as a summary of anything.
Compute MDE = 2.80 * 2100 * sqrt(1/1400 + 1/1400) = 2.80 * 2100 * 0.0378 = 222 compute-hours, 34.7% relative.
Invert to the required N for a 32-hour effect: 4 * (5883 / 32)^2 = about 135,000 accounts, 48x the population.
Recompute under CUPED at rho = 0.80: variance multiplier 0.36, SE multiplier 0.60, MDE 133 compute-hours, 20.8% relative.
Recompute on the binary variant at base rate 0.46 and write the recommendation as one decision with one operating point.

Expected result: MDE is about 222 compute-hours (34.7% relative) as specified and about 133 (20.8%) with CUPED at rho = 0.80. A 5% lift would require roughly 135,000 accounts, about 48 times the eligible population, so the test is declined on this metric and re-specified on a bounded outcome where the MDE falls to about 5.3 percentage points on a 0.46 base.

Verify it yourself

sqrt(1/1400 + 1/1400) = 0.0378, so the MDE must land within a few percent of 2.80 * 2100 * 0.0378 = 222. A materially different number means a one-sample or one-sided formula was used.
Required N scales with the inverse square of the effect size: halving the target MDE must quadruple N. If it does not, the formula is wrong.
The CUPED variance multiplier is 1 - rho^2, never 1 - rho. At rho = 0.80 the standard error falls by 40%, not by 20%.

Follow-up

The feature only applies to the 310 accounts on committed_consumption contracts. What is the MDE on that subset, and does it change the recommendation?
Suppose the primary outcome becomes a binary indicator, account exceeded 100 compute-hours in the month, at a base rate of 0.46. What is the MDE now, and why did it improve so much?
What would make you stop this test early, and how do you write that rule so it is not peeking?

Blended gross margin fell while every segment improved

Medium · mix shift · simpsons paradox · as-of join

Blended gross margin across paying accounts fell from 71 percent to 66 percent over two quarters, yet margin improved inside every plan_tier and every deployment_model. You have fct_usage_daily (account_id, sku_code, usage_date, net_amount_cents, cogs_cents), fct_subscription_period (account_id, plan_tier, term_start_date, term_end_date) and dim_account (account_id, deployment_model, is_internal, effective_from, effective_to). Produce an exact decomposition of the five-point move into mix, rate and interaction terms, and name the segment shift that carries it.

Approach

Write blended margin as a revenue-weighted average of segment margins, m = sum over i of w_i * m_i, where w_i is the segment's share of net_amount_cents. Margin is a ratio of sums, so only revenue weights reproduce the blended figure.
Apply the exact three-term decomposition: delta_m = sum(delta_w_i * m_i0) + sum(w_i0 * delta_m_i) + sum(delta_w_i * delta_m_i), read as mix, rate and interaction. It is exact by construction, so the three terms must sum to the observed change with no residual.
Build segments from as-of joins: take the fct_subscription_period version whose term brackets each usage_date and the dim_account version whose effective_from and effective_to bracket it, rather than joining on is_current and backdating today's attributes over last year's usage.
Run the decomposition on more than one segmentation, at minimum plan_tier, deployment_model and sku_code, since the mix that actually moved may not be the one anybody already suspected.
Once the driving segment is identified, determine whether the shift came from acquisition landing new accounts in a lower-margin segment or from existing accounts migrating, because those have different owners and different remedies.

Follow-up

The interaction term is sometimes large. What does it mean operationally, and when would you prefer a log-mean decomposition that distributes it across the other two terms?
Revenue-weighted average margin is one construction. What changes if leadership wants margin weighted by account count, and which question does each version answer?

Suggested practice checklist

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.

Day 1 — Metric 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.

Day 2 — Diagnosing 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.

Day 3 — Should 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.

Day 4 — The 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.

Day 5 — Technical 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.

Day 6 — Turning 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.

Day 7 — Mock 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.

Getting an analysis acted on

Work that nobody used is a common and unflattering pattern in data careers, and interviewers probe for it. Have a story about an analysis that changed a decision, and be specific about how you got it in front of the person who could act. Also have one about work that went nowhere, with your reading of why.

Explain a time when you had to present a complex technical solution to a non-technical stakeholder or client.

Medium · behavioural and stakeholder questions

Approach

Name the disagreement or constraint, and how you resolved it with evidence.
Pick a story where you drove the decision, not one where you observed it.
Quantify the outcome, including what you would not claim credit for.

Follow-up

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

How do you handle missing or corrupted data in a batch processing pipeline without halting the execution flow?

Medium · behavioural and stakeholder questions

Approach

Quantify the outcome, including what you would not claim credit for.
Name the disagreement or constraint, and how you resolved it with evidence.
Pick a story where you drove the decision, not one where you observed it.

Follow-up

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

Scope an open-ended request to predict account churn

Medium · scoping · churn modelling · operating point

A customer success director asks for a list of accounts about to churn. You know only that the team has six people and that contracts are annual. Available data is fct_subscription_period, fct_usage_daily, fct_api_request, fct_support_ticket and dim_account. Before writing any code, produce the questions you need answered, a proposed definition of about to churn, and the shape of the artefact you would hand back, including the operating point that turns a score into a decision.

Approach

The interviewer is probing whether you convert a vague request into a decision with a capacity constraint attached. A candidate who starts talking about model families has already failed the exercise.
Pin the event and the horizon first. Churn is only possible at term_end_date, so the population is accounts renewing in the next 60 to 90 days, not the whole base. Ask explicitly whether contraction and downgrade count as churn or only full non-renewal, because the three have different base rates and different interventions.
Pin the action and the capacity. Six people times a realistic number of meaningful interventions per week gives k, and k is what the list is ranked to. Evaluate on precision at k rather than a global AUC over accounts that will never be contacted.
Audit leakage before choosing features. Every feature needs a timestamp proving it existed before the prediction date. A downgrade amendment, a churn reason code, and a ticket opened after the renewal conversation started are all leaks that will make the offline number look excellent and the live list useless.
Ask for the counterfactual now rather than later. Coverage is assigned deliberately, so without a held-out slice agreed at the start the intervention can never be evaluated, and you will be asked for its impact in nine months regardless.
Propose the smallest artefact that closes the loop: a weekly ranked list sized to capacity with two or three inspectable reasons per row, plus a stated policy for accounts below the line.

Follow-up

The director insists all accounts are in scope, not only those renewing soon. How do you answer without simply refusing?
Historical non-renewals number about 30 a year. At what point do you tell them a model is the wrong tool and a rules list is better?
Which candidate features would you drop purely because you cannot date them?