Interview guides · DHL

DHL Data Scientist Interview Questions & Guide 2026

Data scientist interview prep for DHL: original practice on experiment design, SQL and metric reasoning, with worked solutions.

The 60-second brief

As a Data Scientist at DHL, you are positioned at the intersection of global logistics complexity and cutting-edge artificial intelligence. Your work is critical to optimizing the massive, real-time supply chain networks that keep goods moving across the globe. By leveraging data, you directly impact operational efficiency, cost-reduction strategies, and the design of intelligent systems that predict shipping patterns and manage capacity.

Ask early whether the loop includes an asynchronous take-home or a timed live case, because the two are graded on different things. A take-home is read as an artifact: the question you decided to answer, what you did about missing or malformed records, and a conclusion stated plainly enough for someone to act on. A reviewer who cannot rerun your notebook discounts the result whatever score is printed in it. Hold to the stated time box and write down what you would have done with more of it, since the follow-up round is usually a live defence of the same work.

Focus areas

Separate true demand from censored, stocked-out sales
Size safety stock under variable lead times
Trace an on-time miss to one node

Overview

This role is both intellectually demanding and highly strategic. You will be tasked with transforming raw, high-volume logistics data into actionable insights that inform everything from delivery route optimization to complex pricing models. Whether you are working within the DHL Data & AI division or supporting specific logistics business units, you will be expected to bridge the gap between technical machine learning solutions and tangible business outcomes.

Expect to work in a collaborative, international environment where your ability to communicate complex findings to non-technical stakeholders is just as vital as your coding prowess. Success at requires a blend of rigorous analytical thinking, a pragmatic approach to problem-solving, and the resilience to navigate the scale and ambiguity inherent in global logistics.

The DHL interview process

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

Initial Screening

Whoever runs this call is usually not a practitioner. They take notes, and a hiring manager skims those notes later, so the real question is whether your work survives being written down by someone outside the field. Test every project sentence against that: could a non-specialist repeat it correctly without knowing what a propensity score is? Carry a plain-language version of each project and one reason you want this particular role that you could not copy onto another application. Vagueness at this stage reads as inexperience, even when the underlying work was genuinely deep.

What to demonstrate

Whether a non-specialist can restate your projects accurately, since their paraphrase is what reaches the hiring manager
Whether your reason for wanting the role points at the work itself rather than the company's reputation
Whether your language signals the level being screened for: what you decided yourself versus what you were handed

How to prepare

Write a two-sentence, jargon-free version of each major project: the question nobody could answer, and the decision your work changed. Read it to someone outside data and have them repeat it back
Point your 'why this role' answer at something concrete in the job description or the product surface you would be working on, and keep it to two sentences
Have two questions ready about measurement: which metric the team is held to, and who acts on an analysis once it lands

Technical Deep-Dive

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

Case Study Round

A case round is decided by whether you leave the interviewer with a recommendation, not by how much analysis you narrate on the way there. The prompt is open on purpose, so the first job is to convert it into a decision someone could act on: ask what would be done differently depending on the answer. From there name the quantity that would settle it, state the assumptions you need, and commit. Candidates who cover more ground than anyone expected and still end on "it depends" score below candidates who scoped narrowly and said what they would do.

What to demonstrate

Whether the version of the question you choose to answer is genuinely narrower than the prompt and still worth answering
Whether the recommendation arrives as an action with a number attached, rather than as a summary of what you looked at
Whether assumptions are stated at the moment you rely on them, instead of collected into a disclaimer at the end
Whether you notice when a branch you are exploring would not change the decision either way

How to prepare

Take six open prompts and write only the scoping move for each: the one-sentence question you would actually answer and the decision it feeds. Give yourself three minutes per prompt and stop there.
Put a five-minute warning into every practice case and force a closing statement that names the action, the result that would justify it, and the result that would reverse it.
Record one case and count how long you talked before naming a measurable quantity. Past roughly five minutes, what you are calling scoping is narration.

Where candidates lose points

Attributing demand variability to the node where it is observed

Order variance amplifies as it moves upstream: batching to a truckload, minimum order quantities, forecast-driven ordering and promotional pull all convert smooth end demand into lumpy replenishment orders, so a plant can see a coefficient of variation several times that of the underlying consumption. Diagnosing the plant's schedule instability as a plant problem then produces interventions that cannot work, because the generating process sits one or two echelons downstream. Measure the bullwhip ratio explicitly (variance of orders placed by a node over variance of demand it received) at each echelon, and fix the ordering rule at the node where the ratio jumps rather than the node where the pain is felt.

Computing average inventory from period-end snapshots

Shipments cluster before period close, so the month-end on-hand position is systematically the lowest point of the month; turns computed against it are biased high and days of supply biased low, frequently by ten to twenty percent, and the bias grows precisely when close-period push is strongest. The same shape of error appears when a daily snapshot is joined to shipment events on date equality: a SKU with several legs on one day fans the snapshot out and multiplies the valued inventory. Average across every daily snapshot in the window for the denominator, and join snapshots to events with an explicit as-of condition and a row-count check at each grain before aggregating.

Averaging per-user rates to produce a population rate

Decide which quantity you want: the mean of per-user ratios and the ratio of summed numerator to summed denominator are different estimands, and heavy users dominate one but not the other. For a ratio metric, aggregate numerator and denominator separately and use the delta method for its variance.

Reading experiment results before checking the arm split

Compare observed arm counts against the intended allocation ratio, not an assumed even split, and set the alarm far below the conventional 0.05: at 0.05 roughly one healthy experiment in twenty trips it, which is why sample-ratio checks usually run at p < 0.001 or stricter. The test's power scales with sample size, so it misses a real diversion on a small experiment and fires on an imbalance too small to move the estimate on a very large one. A flag means go find the assignment or logging fault before reading any outcome, not report a mismatch.

Questions & practice

How would you set up an A/B test for a new dynamic pricing algorithm?

Medium · machine learning and modelling

Approach

Say how the offline result would be validated online before it is trusted.
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.

Follow-up

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

Simulate service levels under a variable inbound lead time

Medium · safety stock · monte carlo · service level

Daily demand for one SKU at one node is Poisson with mean 40. Replenishment lead time is 7 days with probability 0.7, 10 days with 0.2, and 14 days with 0.1, drawn independently per order. The policy is continuous review: order Q = 400 whenever the inventory position drops to the reorder point. Size safety stock two ways for a nominal 95 percent target, once as z times sigma_D times sqrt(mean lead time) and once with the variable-lead-time term. Simulate both and report achieved cycle service level and achieved unit fill rate.

Approach

Compute both safety stock numbers analytically before simulating, so the simulation is a check on arithmetic you already understand rather than the source of the answer.
Track inventory position (on hand plus on order) for the reorder trigger and on hand separately for the stockout test. Triggering off on hand alone re-orders while stock is already in transit and silently changes the policy you are measuring.
Define the two outcomes precisely and separately: cycle service level is the fraction of replenishment cycles containing at least one unit of unmet demand, while fill rate is total units shipped from stock over total units demanded. They answer different questions and z targets only the first.
Before running anything, bound the achievable service level by hand from the lead-time mixture. Each lead-time branch is a separate Poisson demand, so cycle service level is the probability-weighted sum of P(Poisson(40 x L) <= reorder point) across the three branches, and a reorder point that cannot cover the 14-day branch caps the policy at 0.9 no matter how the simulation is tuned.
Run enough cycles that Monte Carlo error is small against the gap you are trying to see. The standard error of a service level is sqrt(p*(1-p)/n_cycles), about 0.005 at p = 0.7 with 10,000 cycles, so the 20-point gap between the two policies is unambiguous and a 1-point one is not.
Let unmet demand be lost rather than backordered, state that choice, and note that switching to full backordering raises measured fill rate for the same policy.

Worked solution (35 min)

Compute mean lead time 8.3 days, variance of lead time 5.01, sigma_D = sqrt(40) = 6.32, and mean demand 40 per day.
Naive safety stock: 1.645 * 6.32 * sqrt(8.3) = 30 units, giving a reorder point of 332 + 30 = 362. Correct safety stock: 1.645 * sqrt(8.3 * 40 + 40^2 * 5.01) = 1.645 * 91.4 = 150 units, giving a reorder point of 482.
Write one simulation loop parameterised by safety stock: draw daily Poisson demand, decrement on hand, record shortfall, trigger an order of 400 when position hits the reorder point, and schedule the receipt after a sampled lead time.
Run 10,000 cycles per policy with a burn-in of at least two lead times discarded so the starting position does not flatter the first cycles.
Report achieved cycle service level, achieved fill rate and the average on-hand position for each of the two safety stock settings.

Expected result: Naive safety stock is 30 units and the variable-lead-time figure is 150 units, a factor of five. At Q = 400 with lost sales, the naive policy achieves a cycle service level of about 0.70 and the correct policy about 0.90, so the naive policy misses the nominal 0.95 by roughly 25 points and the correct one by 5. The naive ceiling is arithmetic, not simulation noise: with a reorder point of 362 the 7-day branch is essentially always covered, the 10-day branch survives with probability P(Poisson(400) <= 362) = 0.029, and the 14-day branch never does, so cycle service level cannot exceed 0.7 + 0.20.029 + 0.10 = 0.706, and reorder-point undershoot of about 20 units under daily demand arrivals pulls it to roughly 0.70. The correct policy tops out near 0.90 for the same structural reason: a reorder point of 482 covers the 10-day branch almost completely (P = 0.99997) but not the 14-day one (P = 0.0004), so that 0.1 of probability mass is near-pure loss. Fill rate is higher than cycle service level under both, about 0.92 against 0.70 and about 0.98 against 0.90, because expected units short per cycle (roughly 33 and 10) are small against Q = 400.

Verify it yourself

Set the lead time to a constant 8.3 days and confirm the two formulas return an identical safety stock, since the lead-time variance term vanishes.
Compare the naive policy's simulated cycle service level against its analytic ceiling, 0.7 + 0.2P(Poisson(400) <= 362) + 0.1P(Poisson(560) <= 362) = 0.706 before undershoot. A simulated figure above that is a bug, usually triggering off on hand instead of inventory position, and three seeds should land a little below it and agree within about one point, consistent with a standard error of about 0.005 at 10,000 cycles.
Confirm fill rate exceeds cycle service level at Q = 400 under both policies, and treat that ordering as a consequence of Q rather than a law: expected units short per cycle depends on the reorder point and the lead-time demand distribution, not on Q, so holding the reorder point fixed and halving Q leaves cycle service level roughly unchanged while fill rate falls by about the same shortfall over a smaller denominator.
Confirm simulated demand over the lead time has mean near 332 units and standard deviation near 91.4, which is exactly the quantity the correct formula multiplies by z.

Follow-up

With Q = 100 against lead-time demand of 332 units, three or four orders are open at once. What happens to the meaning of a replenishment cycle, and to each of the two service measures?
Lead time and demand are correlated because the supplier is capacity constrained in peak weeks. What does that do to the formula you used?
How would you convert the extra 120 units of safety stock into an annual cost the business can weigh against the service gain?

As-of join snapshots to order lines without merge_asof

Medium · as-of join · fan-out · searchsorted

You have fct_order_line (order_line_id, sku_id, ship_from_location_id, actual_ship_at_utc, shipped_qty, shipment_id) and fct_inventory_daily (inventory_date, sku_id, location_id, available_qty). For every shipped line, attach available_qty from the latest snapshot strictly before the ship date for that exact sku and node, plus the shipment's delivery timestamp from fct_shipment_leg (leg_id, shipment_id, delivered_at_utc, leg_status). The output must have exactly one row per input order line. pandas.merge_asof is not available; implement the as-of match yourself.

Approach

Collapse fct_shipment_leg to one row per shipment_id before joining anything. A shipment is one or more legs, so a direct line-to-leg join multiplies every line by its leg count and quietly inflates every sum computed afterwards.
Define the shipment-level delivery as the maximum delivered_at_utc across its legs, and set it null if any leg is not delivered, since the customer has the order only when the last leg lands.
Sort the snapshot frame by (sku_id, location_id, inventory_date) and the line frame by the same key plus ship date, then use np.searchsorted with side='left' on the sorted dates within each key group to find the insertion point; the matching snapshot is the row at position minus one.
Guard the boundary explicitly: an insertion point of 0 means no snapshot exists before that ship date, and the result must be null rather than the first available future snapshot. Count those lines and report them.
Assert the output row count equals the input line count immediately after each merge, before any aggregation. Row-count drift is the only symptom a silent fan-out produces, and it is invisible once the frame has been grouped.

Follow-up

The snapshot is taken at the node's local end of day and actual_ship_at_utc is UTC. Where does that break the strictly-before rule and what do you need from dim_location to fix it?
How would you extend this to allocate each leg's linehaul cost back to the order lines it carried?
What changes if a line ships in two partial shipments?

Join these two tables and identify the top 10% of customers by shipment volume.

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.
Check whether any join is one-to-many before aggregating, or the sums inflate.

Follow-up

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

How would you handle missing values in a dataset containing millions of shipment records?

Medium · sql and data manipulation

Approach

Compute rates by summing numerator and denominator separately, never by averaging rates.
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.

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?

Write a query using SQL window functions to calculate the rolling average of shipping delays per region over the last 30 days.

Medium · sql and data manipulation

Approach

Compute rates by summing numerator and denominator separately, never by averaging rates.
State the window function and its partition and ordering out loud before writing it.
Check whether any join is one-to-many before aggregating, or the sums inflate.

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?

Landed cost per delivered unit by lane without fan-out

Medium · fan-out · many-to-many · cost per unit

fct_shipment_leg holds one row per leg: shipment_id, leg_seq, origin_location_id, destination_location_id (NULL when the stop is a customer address rather than a network node), direction, mode, shipped_units, stop_count, leg_status, and linehaul_cost_cents, fuel_surcharge_cents, accessorial_cost_cents and expedite_premium_cents. fct_order_line carries shipment_id, NULL until shipped. Compute monthly landed cost per delivered unit by lane for direction in ('outbound','interfacility'), keying the lane on origin and destination with customer stops bucketed. Do not route through fct_order_line to get units. State the multi-stop allocation basis and what shipped_units means on a leg.

Approach

Keep the entire computation at leg grain. Numerator is the sum of the four cost columns, denominator is SUM(shipped_units), both over legs with leg_status = 'delivered' and the two directions. The order table is not needed and joining it can only damage the number.
Bucket destination_location_id IS NULL into an explicit customer_stop lane key with COALESCE, because a NULL never equals anything and will otherwise drop the row from a join or split it silently in a GROUP BY comparison.
Say out loud whether shipped_units is units on board or units dropped at that stop. Under units on board, summing the legs of a multi-stop load counts a unit once per leg it rides and understates cost per unit; under units dropped, the leg sum is a true shipment total.
Declare whether multi-stop cost is allocated by weight or by cube, since dense and bulky freight rank lanes in opposite orders under the two bases and the lane league table is not readable without it.
Roll months up by re-summing cost and units and dividing once. Averaging leg-level cost per unit weights a one-pallet LTL leg the same as a full truckload.

Worked solution (25 min)

Filter fct_shipment_leg to leg_status = 'delivered', direction IN ('outbound','interfacility'), and the month range.
Build lane_key from origin_location_id and COALESCE(destination_location_id::text, 'customer_stop').
Aggregate per month and lane_key: total_cost_cents as the sum of the four columns, total_units as SUM(shipped_units), leg_count as COUNT(*).
Divide once: cost_per_unit_cents = total_cost_cents / NULLIF(total_units, 0).
Cross-check the grand total against an unaggregated SUM over the same filter, and record the allocation basis and shipped_units semantics in the output header.

Expected result: One row per month and lane with total_cost_cents, total_units, leg_count and cost_per_unit_cents, computed only from fct_shipment_leg, where leg_count equals a plain COUNT(*) over the identical filter.

Verify it yourself

SUM(shipped_units) over the leg set equals a standalone aggregate with no joins. Routing through fct_order_line makes it larger, which is the fan-out showing itself.
Lane cost totals sum to the unfiltered month total for those directions and statuses.
Legs with destination_location_id NULL appear in the output under the customer bucket rather than going missing.

Follow-up

Cost per unit rose 8 percent while contracted linehaul rates were flat. Which columns do you cut first, and what does a rise in accessorials rather than linehaul point to?
How do you separate a rate change from a mix change across mode and service_level?
If cost genuinely had to be attributed to an order line, how would you allocate a leg's cost, and what does your rule do to a line that shipped one heavy item on an otherwise full truck?

If the "on-time delivery" metric drops by 5% overnight, how would you investigate the root cause?

Medium · product sense and metrics

Approach

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.
Decompose the metric into the rates that drive it, and say which one you would check first.

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
Which segment would you cut first, and what would that rule out?

What product metrics would you track for a new automated package-sorting dashboard?

Medium · product sense and metrics

Approach

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

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
Which segment would you cut first, and what would that rule out?

How would you design a metric to measure the efficiency of a last-mile delivery route?

Medium · product sense and metrics

Approach

Restate the decision this analysis has to support, and who acts on the answer.
Decompose the metric into the rates that drive it, and say which one you would check first.
Fix the population and the time window before naming any metric.

Follow-up

How would you detect that the metric is being gamed rather than genuinely improving?
Which segment would you cut first, and what would that rule out?

What are the most common experimentation pitfalls when testing changes in a physical warehouse environment?

Medium · a/b testing and experimentation

Approach

Say whether units interfere with each other, and switch design if they do.
Name the guardrails that would stop a launch even on a positive primary result.
Name the randomisation unit first; it decides the variance and what the test can detect.

Follow-up

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

How do you determine if a result is statistically significant when dealing with highly seasonal shipment data?

Medium · a/b testing and experimentation

Approach

Name the guardrails that would stop a launch even on a positive primary result.
State the primary metric and the minimum effect worth shipping, then size the test.
Say whether units interfere with each other, and switch design if they do.

Follow-up

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

Value a staggered forward-stocking rollout that was never randomised

Hard · difference-in-differences · synthetic control · staggered adoption · spillover

Six forward-stocking locations were opened over five months, chosen because their regions had the worst P90 order cycle time. There is no holdout. You have 24 months of node-month history in fct_order_line and fct_shipment_leg for all 40 nodes, and each new location draws its volume from the regional DC that used to serve the same customers. Estimate the effect on P90 order cycle time and on landed cost per delivered unit, and state the assumption your estimate rests on and how you test it.

Approach

Fix the unit of analysis before choosing an estimator. Volume transfers from the parent regional DC to the new location, so the parent is treated by interference: its mix, lane density and cost per unit all change. Aggregate to the region, meaning new location plus parent, so the transfer is internal, and keep parents out of any comparison group.
Name the selection problem. Sites were picked on the worst pre-period outcome, so mean reversion alone manufactures an apparent improvement. Difference-in-differences does not require matching levels, only that untreated trends would have been parallel; test that with an event-study specification whose lead coefficients should be indistinguishable from zero, and look specifically for a pre-period dip that would have reverted on its own.
Avoid the naive two-way fixed effects estimator. With five distinct opening dates and effects that grow after opening, TWFE uses already-treated regions as controls for later ones and can put negative weight on those comparisons; use a group-time estimator such as Callaway and Sant'Anna or Sun and Abraham with never-treated regions as the comparison group, then aggregate to a dynamic path.
Run synthetic control per treated region as an independent second read: fit donor weights on pre-period P90, volume and lane mix, require pre-period fit error that is small relative to the post-period gap, and obtain inference from placebo-in-space permutations, ranking the treated gap against every donor's placebo gap.
Be honest about inference with six treated units. Cluster-robust standard errors are badly anti-conservative at that count, so use a wild cluster bootstrap with Rademacher weights or randomisation inference over opening dates, and report both outcomes together, since cycle time bought with expedite premium is a different result from cycle time bought with position.

Worked solution (40 min)

Build a region-month panel: P90 order cycle time computed from raw line-level durations per region-month, plus landed cost per delivered unit, plus an opening-date column.
Fit an event study relative to opening month with region and calendar-month fixed effects, and inspect leads from minus six to minus one.
Estimate group-time average treatment effects with never-treated regions as controls, then aggregate to a dynamic path and a single post-period average.
Fit a synthetic control per treated region, compute the treated gap and every donor's placebo gap, and report the rank-based p-value.
Re-run inference with a wild cluster bootstrap at the region level, and present the DiD and synthetic control estimates side by side, flagging disagreement rather than picking the friendlier one.

Expected result: A dynamic treatment path for each outcome, with the identifying assumption stated as parallel trends conditional on the covariates used, supported by flat pre-period leads and a placebo-permutation p-value from the synthetic control. The defensible claim is an effect for these six regions over this window, reported jointly with the cost outcome, not a network-wide projection.

Verify it yourself

Region volume should be conserved across each opening: orders shipped by the new location plus its parent should track the region's prior demand, or volume is leaking from somewhere the model does not include.
Each synthetic control's pre-period RMSPE should be small relative to its post-period gap; a ratio near one means the fit carries no information.
A placebo opening date set six months before the real one should produce no effect.
P90 must be recomputed from line-level durations at every level of aggregation, since percentiles do not average and a rolled-up P90 of P90s is not the P90.

Follow-up

Your event study shows a downward lead coefficient two months before opening. What are the two explanations, and how do you tell them apart?
One region's synthetic control has a pre-period fit error as large as its post-period gap. What do you report for that region, and does it enter the pooled estimate?

Forecast accuracy collapsed at nodes missing snapshot rows

Medium · pipeline failure · wmape · data completeness

WMAPE at lag 7 for one region jumped from 31 percent to 58 percent over three weeks with no model change and no retraining. fct_inventory_daily is specified to carry one row per active sku-location-date even when on_hand_qty is zero, and a release changed how the end-of-day cut-off is derived from dim_location.timezone. Using fct_inventory_daily (inventory_date, sku_id, location_id, demand_qty, forecast_qty_lag7, stockout_flag) and dim_location, decide whether demand changed or the table did, and quantify the contaminated share before anyone retrains.

Approach

Run a completeness check before touching any accuracy number: expected rows per location per day from the active sku-location grid, against actual rows present. If the table is incomplete the accuracy question is not well posed, and this check costs one query.
Separate the two failure modes a cut-off change produces. Rows missing entirely shrink the denominator, and the sku-days that vanish are not a random sample. Rows present but with demand pushed onto the adjacent date create a one-day shift that inflates absolute error on both days while leaving the weekly total intact.
Test the shift hypothesis directly by comparing SUM(demand_qty) per ISO week per node before and after the release. If weekly totals match and only the daily split moved, the model is fine and the grain is broken.
Recompute WMAPE on clean node-days only and report both figures with the excluded row count and the share of demand those rows represent, since the metric definition already requires the excluded count to travel with the number.
Check whether the affected nodes share a timezone that crosses a daylight-saving boundary or sits far from the offset the pipeline assumed. That grouping is the discriminating evidence between a code defect and genuine demand movement.

Follow-up

Weekly totals match and only the daily split moved. Does the ordering decision care? At what replenishment frequency does it stop mattering?
How would you backfill the missing days, and which of them would you refuse to backfill?
What assertion would you add to the snapshot job so this fails loudly instead of surfacing three weeks later as model drift?

Suggested practice checklist

Roughly 90 minutes a night on weekdays with one longer weekend block. The plan deliberately cuts scope rather than compressing everything, on the assumption that finishing one thing a night beats half-starting four.

Day 1 — Fix the scope and set a baseline

Read the role description and write the three things the loop will almost certainly test, then write an explicit not-doing list for everything else and keep it visible all week.
Take one 20-minute SQL prompt and one 10-minute metric question cold, and write the single sentence that says what blocked each attempt, since that sentence is what decides which two topics get the most evenings.
Set the week's one rule: one problem finished to completion every night, including the night you only have 40 minutes.

Deliverable: A one-page scope with an explicit not-doing list and two cold attempts, each carrying one sentence on what blocked it.

Day 2 — One query pattern, written three times

Choose the single pattern most likely to appear (a cohort retention grid, or a funnel counted by user) and write it three times from a blank file rather than editing the previous attempt.
On the third attempt, write the grain of every CTE as a comment before writing its body.
Stop at 90 minutes even if the third version is imperfect, and write the one thing you would fix with another hour.

Deliverable: Three independent versions of the same query plus a note on what changed between them.

Day 3 — Only the statistics you will be asked to defend

Write, in under 200 words, how you would decide whether a difference between two groups is real: the test, its assumptions, and what you would switch to when an assumption fails.
Compute a 95 percent confidence interval for a difference in proportions by hand on realistic numbers, then write in one sentence what changes if the two samples are paired rather than independent.
Write your answer to "what does a p-value mean", check it against a definition, and delete the version that describes it as the probability the hypothesis is true.

Deliverable: A 200-word written answer and one hand-computed interval you can reproduce under pressure.

Day 4 — One case, and the assumptions holding it up

Answer one product case aloud in 20 minutes with a recording running, then listen back with a pen and mark every claim you asserted without saying what it rested on: an assumed user behaviour, an assumed data source, an assumed baseline rate, an assumed grain.
Pick the three assumptions the recommendation actually depends on, write how you would check each one against data, and say which one being wrong would flip the recommendation rather than merely weaken it.
Write the four-step structure you used onto a card small enough to hold in working memory when you are nervous.

Deliverable: One recording, three load-bearing assumptions each with a written check, and a four-step structure card.

Day 5 — Your own work, timed

Write a 90-second version and a four-minute version of your main project, and time both out loud rather than reading them.
Prepare answers to the two follow-ups that always come: what you would do differently, and how you knew it worked.
Put one number in the first sentence and be able to say exactly where that number came from and what it excludes.

Deliverable: Two timed narratives with one defensible number in the opening line.

Day 6 — The one full rehearsal, in a longer weekend block

Run a 60-minute mock covering query work, a case and a behavioural question in a single sitting with no breaks, because sustained attention is the thing evenings have not trained.
Immediately afterwards, and before hearing any feedback, write the three moments you lost the thread.
Spend the rest of the block only on those three moments, and on nothing you merely feel shaky about.

Deliverable: Mock notes naming three failure moments with a specific fix written under each.

Day 7 — Taper

Write the 20-minute warm-up you will actually do on the morning of the interview: one query you can already write from a blank file, one metric you can define out loud, and nothing you have never seen before.
Re-read only your own notes from this week, and open no new material.
Write down the logistics: the tool you will be asked to work in, whether lookups are allowed, and the sentence you will use when you do not know something.

Deliverable: A one-page card holding the case structure, the project numbers, and the logistics.

Rigour against a deadline

Nearly every data role forces a trade between the analysis you want and the one that fits the decision window. Prepare a case where you deliberately shipped something less rigorous, named the weakness to the person relying on it, and said what would change your answer. The naming is the part interviewers listen for.

Describe a situation where you encountered a significant roadblock in a project; how did you overcome it?

Medium · behavioural and stakeholder questions

Approach

Close with what you would do differently, concretely.
Quantify the outcome, including what you would not claim credit for.
State the situation in two sentences and spend the rest on your reasoning.

Follow-up

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

Defend a finding that the expedite program bought nothing

Medium · defending findings · counterfactual reasoning · landed cost

Over two quarters, expedite_premium_cents on outbound legs in fct_shipment_leg rose from 0.4 to 1.3 percent of landed cost, while perfect order rate moved from 91.2 to 91.5 percent, inside the week-to-week spread. The transport lead who sponsored the expedite program disputes the finding in a review in front of his director, arguing that your window contains a port disruption that would have made service worse without the spend. Present the finding, say what you concede on the spot, say what you hold, and name the evidence that would change your conclusion.

Approach

Restate his objection in its strongest form before answering it, because a counterfactual worsening is a legitimate argument and treating it as an excuse ends the conversation.
Separate what you measured from what you claimed: the data show no detectable service gain, not that expedite has no effect, and the difference is the entire argument.
Test his hypothesis with the data you already have rather than defending in the abstract: split legs by exception_code = 'customs_hold' and by lane, and compare expedited against non-expedited legs on the same lanes in the same weeks, since if expedite were holding the line the expedited lanes should show a service gap over comparable non-expedited ones.
Concede what is true: without a holdout you cannot rule out a protective effect, and the pre-period is contaminated by the disruption, so the honest statement is an upper bound on the gain rather than a zero.
Hold the part that survives: the spend is real, it is concentrated in a small set of lanes, and nobody set a decision rule for when a leg gets expedited, which is a controllable problem independent of the counterfactual.
Name the next measurement: a lane-level staggered switch-off with a stated burn-in, and say what it would cost and how long it would take.

Follow-up

He offers to run the switch-off only on his two best lanes. Why is that a problem, and what do you counter with?
His director asks you for a yes or no on cutting the budget today. What do you say?

Prioritise transport, plant and commercial in one week

Medium · prioritisation · stakeholder management · decision value

Three requests land in the same week. Transport wants landed cost per delivered unit split by lane and mode from fct_shipment_leg for a carrier bid that closes in nine days. A plant manager wants OEE on two work centres decomposed into availability, performance and quality from fct_production_run for a capital request due next month. Commercial wants a fill-rate root cause from fct_order_line for an account review on Thursday. You are the only data scientist and have about four working days. Give your order, the rule that produced it, and what you say to the two who wait.

Approach

Rank by the decision each request feeds and by whether your input can still change it, not by seniority or by who asked loudest: a bid closing in nine days is a live decision with a large irreversible spend attached, a capital request due next month has slack, and an account review has a fixed date but a smaller reversible decision.
Check reversibility and blast radius: carrier rates lock for a contract period across every lane, so an error or an absence there is expensive for a year, while the fill-rate story can be revised next week.
Look for the cheap partial that unblocks someone else: the fill-rate cut by node, by short_reason_code and by week is a few hours of work and covers most of what commercial needs on Thursday, so it does not have to wait behind the bid.
Sequence with explicit time boxes: the commercial cut first because it is short and date-locked, the lane and mode cost analysis next with the allocation basis for multi-stop loads stated up front, and the OEE decomposition scheduled into the following week with a date the plant manager can hold you to.
Tell the plant manager directly and early, with a date rather than a maybe, and say what you need from him in the meantime so the delay produces something.
Escalate the trade-off rather than absorbing it silently: your manager should know a capital request slipped a week, because that is a business choice and not yours alone to make.

Follow-up

The plant manager escalates and his director asks you to reorder. What do you do?
Which of the three would you push back on entirely, and what would you offer instead?