Interview guides · AXA

AXA Data Scientist Interview Questions & Guide 2026

Timed data scientist practice for AXA: short SQL and probability sets run end to end, then scored against a rubric.

The 60-second brief

As a Data Scientist at AXA, you occupy a highly strategic and impactful position within one of the world's largest insurance and asset management companies. Data is the lifeblood of modern insurance, and your role is to transform massive, complex datasets into actionable insights that directly influence business decisions. From optimizing risk assessment and fraud detection to automating claims processing and refining customer personalization, your work has a direct line of sight to AXA's bottom-line performance and customer satisfaction.

Learn the economics of the product category before the loop. Marketplaces, subscription products and ad-supported products turn on different core quantities (match rate and liquidity, retention and churn, fill rate and yield) and fail in different characteristic ways.

Focus areas

Read vintage curves, not blended portfolio averages
Set fraud thresholds by expected cost
Report only matured cohorts for loss metrics

Overview

The work of a Data Scientist at AXA is uniquely challenging due to the scale and regulatory complexity of the insurance industry. You will work with diverse data types, including telematics, claims histories, demographic data, and unstructured text from customer interactions. Unlike tech-first startups, data science in insurance requires a sophisticated balance of cutting-edge machine learning and traditional actuarial rigor, ensuring that models are both highly predictive and fully compliant with strict financial regulations.

This position offers an exciting opportunity to work on high-stakes problems with global reach. Whether you are building predictive models for regional business units or collaborating on global platform initiatives, you will be part of a team that values innovation, structured problem-solving, and cross-functional collaboration. To succeed, you must not only possess deep technical and statistical expertise but also the business acumen to translate complex equations into tangible value for stakeholders.

The AXA interview process

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

Initial Screening Call

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 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

Deep-Dive Interviews

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.

Conversations with Senior Directors

Because the format is not fixed, prepare the reasoning rather than the ritual. Nearly every version of this round draws on the same underlying material: a design you can defend, a metric you can define exactly, an analysis whose assumptions you can state out loud. Only the wrapper changes, whether that is a take-home, a live case, a deep dive on past work, or a rough estimate on a whiteboard. Answers rehearsed to fit one shape stall the moment the shape differs. Practise naming the assumption behind a number, then saying how much the conclusion moves if that assumption is wrong.

What to demonstrate

Whether your justification for a method survives the question 'why not the simpler thing', including when the simpler thing would have worked
Precision under pressure: what exactly counts as an active user, a conversion or a success, over what window, with what exclusions
Whether you carry an argument through to a recommendation instead of stopping at a list of tradeoffs

How to prepare

For each project you plan to mention, write the metric definition in one sentence: numerator, denominator, time window, exclusions. Say it out loud once, because vagueness shows up in speech before it shows up on paper.
Rehearse the same project at three lengths: two minutes, ten minutes, and a deep dive on one technical decision. Cutting live is harder than it sounds.
For your headline result, write down what would have had to be true for it to be wrong, and how you ruled that out.

Where candidates lose points

Reading the most recent months of fraud and dispute rates as final

Consumer dispute rights commonly run around 120 days from the transaction or expected delivery date, and several reason codes run considerably longer, so the disputes belonging to a recent transaction month have simply not been filed yet. Any chart attributed by transaction date therefore slopes down at the right edge regardless of what is happening. The fix is to report only matured cohorts, or to apply development factors estimated from completed months and to show the estimate as an estimate.

Averaging delinquency across a book that is growing

A loan three months old cannot be 90 days past due, so a portfolio with many recent originations reports a low blended 90+ rate purely from age mix. The blended rate falls fastest exactly when originations grow fastest, which is precisely when credit quality most needs watching, so the metric moves in the reassuring direction during the riskiest period. Only comparisons at equal months on book are valid, which is what a vintage or roll-rate view enforces.

Never asking what decision the analysis will inform

Open with who makes the decision, what the options are, and by when. The answer determines the precision you need, the segments worth cutting, and whether an observational read suffices or an experiment is required.

Solving silently instead of narrating the reasoning

Say which branch you are taking and why you chose it over the alternative, for example checking the denominator first because it changes what the comparison means. A correct answer that arrives with no visible path scores below a rigorous one that needed a hint.

Questions & practice

Explain the mathematical difference between L1 (Lasso) and L2 (Ridge) regularization. When would you use each?

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.
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?

Describe how a Gradient Boosting machine works. How does it differ from a Random Forest?

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.
Set a baseline first, so any model has something honest to beat.

Follow-up

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

Simulate false alarms in a merchant chargeback monitoring rule

Medium · simulation · rare events · monitoring thresholds

Baseline matured first-chargeback rate is 12 per 10,000 settled transactions. A monitoring rule alerts when a merchant's observed monthly rate exceeds twice baseline. For monthly settled transaction counts of 500, 2,000, 10,000 and 50,000, simulate the false-alarm probability per merchant-month under the baseline, and the power to detect a merchant whose true rate is 30 per 10,000. Then, for a portfolio of 4,000 merchants split 60, 25, 10 and 5 percent across those four counts, give the expected number of false alarms per month.

Approach

Recognise the rule is a threshold on an integer count, not on a continuous rate. At n = 500, twice baseline is 24 per 10,000, so the first observable value above it is 2 chargebacks, or 40 per 10,000. Derive the trigger count for every n before simulating anything.
Draw binomial counts with numpy at p = 0.0012 and take the share at or above the trigger for the false-alarm rate, then repeat at p = 0.0030 for power. Use at least 200,000 draws per cell so a probability near 0.001 has a usable standard error.
Cross-check every simulated cell against the Poisson approximation with lambda = n*p, which is tight here because p is tiny. A mismatch almost always means the trigger count is off by one.
Weight the per-merchant false-alarm probabilities by the portfolio mix, and report the share of expected alerts contributed by each size band rather than only the total.
Close on the operating consequence: a fixed multiplicative threshold is not a constant false-alarm rate across merchant sizes, so either the threshold scales with n or small merchants need a minimum volume before the rule applies.

Worked solution (30 min)

For each n, compute trigger = floor(2 * 0.0012 * n) + 1 and print the four values before simulating.
Simulate 200,000 binomial draws per n at p = 0.0012 and take the share at or above the trigger.
Repeat at p = 0.0030 and record power for the same triggers.
Compute the Poisson tail 1 - CDF(trigger - 1, lambda = n*p) for both p values and confirm agreement within Monte Carlo error.
Multiply the false-alarm probabilities by 2400, 1000, 400 and 200 merchants and sum.

Expected result: Trigger counts of 2, 5, 25 and 121. False-alarm probability roughly 12 percent at n = 500, roughly 10 percent at n = 2,000, under 0.1 percent at n = 10,000 and effectively zero at n = 50,000. Power against 30 per 10,000 roughly 44, 72, 85 and above 99 percent. Expected false alarms about 390 per month, with over 99 percent of them coming from the two smallest bands.

Verify it yourself

The four trigger counts are derived, not guessed: each is the smallest integer strictly above 0.0024n, giving 2, 5, 25 and 121. Derive them in exact or integer arithmetic, because 20.0012*10000 evaluates to 23.999999999999996 in binary floating point and floor-plus-one silently returns 24.
Simulated probabilities match the Poisson tail at lambda = n*p within about two Monte Carlo standard errors.
Power exceeds the false-alarm rate at every n; if it does not, the null and alternative rates are swapped.
Doubling the draw count moves each probability by less than its own simulated standard error.

Follow-up

How would you set a threshold that holds the false-alarm rate roughly constant across merchant size?
The rule reads the transaction month, but disputes arrive for up to 120 days afterwards. What does that do to the alert and how would you fix it?
What does a month of these false alarms cost, and how would you decide whether it is worth paying?

Explain how you would design a data pipeline to ingest, clean, and store daily telemetry data from connected cars.

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.
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?

Walk me through a logical puzzle: how would you estimate the number of car insurance policies active in a major city without direct access to registry databases?

Medium · sql and data manipulation

Approach

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

Follow-up

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

How would you optimize a slow-running SQL query that joins multiple massive transaction tables?

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.
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 does the query change if the join becomes one-to-many?

Write a Python function to clean and parse unstructured text data from a claims report.

Medium · sql and data manipulation

Approach

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

Follow-up

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

Collapse merchant retry chains before measuring authorization approval

Medium · window functions · gaps and islands · sessionisation

Merchant retry logic turns one declined purchase into several rows in fct_payment_authorization. Collapse attempts into chains: inside a partition of (card_token_id, merchant_id, amount_minor), consecutive rows whose requested_at gap is 15 minutes or less belong to the same chain. Return one row per chain with chain_started_at, attempt_count, amount_minor, channel, and chain_outcome set to 'approved' when any attempt in the chain was approved. Then report a 7-day rolling dollar-weighted approval rate by channel over chains rather than attempts. Exclude is_reversal rows.

Approach

In a CTE, take LAG(requested_at) OVER (PARTITION BY card_token_id, merchant_id, amount_minor ORDER BY requested_at) and set is_new_chain when the previous value is null or the gap exceeds 15 minutes.
Turn the flag into a chain identifier with SUM(is_new_chain::int) OVER (same partition, same order, rows unbounded preceding to current row), which is the standard gaps-and-islands construction.
Aggregate to one row per chain: MIN(requested_at), COUNT(*), and BOOL_OR(auth_result = 'approved') as the chain outcome, because the purchase succeeded if any attempt in the chain did.
Roll the chains to a daily grain per channel, then apply the rolling window as PARTITION BY channel ORDER BY chain_date RANGE BETWEEN INTERVAL '6 days' PRECEDING AND CURRENT ROW. An interval-offset RANGE frame is value-based: it admits every row in the partition whose chain_date falls in the closed window [D - 6 days, D], so a channel with no chains on some dates still gets the right seven-day sums — sparsity is exactly what this construct tolerates. ROWS BETWEEN 6 PRECEDING AND CURRENT ROW is the one that needs a dense grain, because it counts rows and on a sparse grain reaches back further than seven days.
Left-join a dense date spine only if the output needs a row on days with no chains. That is a presentation requirement, not a correctness one, and it leaves every rolling sum unchanged. The RANGE frame's preconditions are that chain_date is a date or timestamp (an interval offset needs an ordering column it can be added to) and that the aggregation left one row per (channel, chain_date).
Divide approved chain value by total chain value inside the rolling frame, never the average of daily rates, since averaging rates weights a quiet day the same as a heavy one.

Worked solution (40 min)

CTE 1: filter out is_reversal rows, then compute prev_requested_at with LAG and the is_new_chain flag.
CTE 2: build chain_id with a running SUM of the flag over the same partition and order.
CTE 3: group by partition keys plus chain_id to emit chain_started_at, attempt_count, amount_minor, channel and chain_outcome.
CTE 4: aggregate chains to (chain_date, channel) with approved value and total value. Add the date spine here only if the report must show empty days.
Final SELECT: two rolling SUMs over PARTITION BY channel ORDER BY chain_date RANGE BETWEEN INTERVAL '6 days' PRECEDING AND CURRENT ROW, divided at the end.

Expected result: A rolling series with one row per date and channel, where the chain count is strictly below the attempt count on ecommerce and recurring, and the chain-level approval rate is at or above the attempt-level rate for every channel, since collapsing removes failed attempts that later succeeded.

Verify it yourself

SUM(attempt_count) over all chains equals the filtered attempt row count exactly; if it does not, the chain identifier is not partitioned the same way as the LAG.
Every chain has exactly one row whose is_new_chain flag is true, so COUNT(DISTINCT chain_id) equals the number of flagged rows.
A chain whose attempt_count is 1 must have chain_outcome equal to that single row's approved status.
The rolling denominator on day D equals the sum of the daily denominators over every date present between D-6 and D inclusive — seven terms on a dense grain, fewer on a sparse one, and the same total either way.
Adding or removing the date spine leaves every rolling sum identical, because spine rows contribute zero. Swapping RANGE for ROWS changes them on any channel whose grain has a gap, which is the direct test of which construct tolerates sparsity.

Follow-up

This definition chains off the previous attempt, so a chain can span an hour if attempts arrive every 14 minutes. When is that right, and when would you anchor the window to the first attempt instead?
Two genuine purchases of the same amount at the same merchant, four minutes apart, collapse into one chain. How large is that error and how would you bound it?
A channel goes dark for three days mid-window. Which of the two frame types would silently widen, and what would the resulting rate look like?
What changes if the merchant retries with a slightly different amount to dodge a velocity rule?

Price a false decline when the label does not exist

Hard · proxy metrics · selection bias · randomised holdout

Your expected-cost threshold needs a figure for what a false decline costs beyond the margin on the blocked transaction. fct_payment_authorization records the decline, nothing records what the customer did next elsewhere, and declined transactions never produce a fraud outcome, so neither side of the error is directly observable. Propose the measurement: the proxy you would build from the tables you have, the design that yields an unbiased estimate for at least part of the score range, the bias in each, and the single sentence you would attach to the number when it reaches a pricing decision.

Approach

Write down what is unobservable and why. The counterfactual spend of a customer who was not declined, and the fraud label on any transaction the rule blocked, are both missing because of the decision itself. Missingness that depends on the decision is not fixed by matching on observed covariates.
Build the observational proxy anyway and be specific. For customers receiving a first risk-rule decline in a window, compare settled volume and active status over the following 30 and 90 days against customers matched on pre-period settled volume, tenure, segment and channel mix who attempted a comparable transaction and were approved. The bias runs one way: matching conditions on having attempted something that scored near the cutoff, and part of the declined group are genuine fraudsters whose disappearance is a saving rather than a loss, so the estimate overstates the damage.
Buy one unbiased local estimate. Hold a small random share of authorizations inside a defined risk_score band out of the decline rule and approve them, sizing the sample in advance from the expected fraud rate in that band so the cost of the experiment is known before it runs. The result is unbiased for that band only, and it is simultaneously the only source of fraud labels in the declined region.
Bound the extrapolation instead of hiding it. Run the holdout in two or three adjacent bands and report the spread. If the effect is flat across bands a constant is defensible; if it is steep, quote band-specific figures and decline to supply a single number.
Handle the window. Attrition after a decline can resolve over months, so a 90-day window truncates it and the randomised estimate is a lower bound on long-run damage at the same time as the observational version is an upper bound. Stating both directions is what makes the number safe to use.
Write the sentence that travels with the number: what it is (an estimate from a randomised holdout in one score band over a 90-day window), what it is not (a measurement anywhere else on the score range), and which way it is likely to be wrong.

Worked solution (40 min)

Define the decline cohort and the matched comparison cohort precisely, including matching variables and the pre-period window, and produce the 30-day and 90-day settled-volume difference.
Decompose the declined cohort into customers who never transact again and customers who transact less, since fraudsters concentrate in the first group and that split tells you how much of the estimate is contamination.
Write the holdout design: score band, sample share, expected fraud rate in band, expected cost of running it, and the run length needed to detect an effect large enough to change the cutoff.
Recompute p* = C_FP / (C_FP + C_FN) at the top and bottom of your estimated cost range and state whether the range changes the cutoff you would set.
Write the one-sentence caveat that will be quoted alongside the number in the pricing decision.

Expected result: Two numbers with opposite biases, an observational estimate labelled as an upper bound and a randomised in-band estimate labelled unbiased for that band only, plus a sensitivity table showing how p* moves across the range and an explicit statement of whether the residual uncertainty changes the decision.

Verify it yourself

If the cost range moves p* by less than the granularity at which the rule can actually be set, further measurement is not worth buying and that is the finding.
The observational estimate should exceed the randomised one in magnitude. If it does not, the matching is capturing something other than the effect of the decline and needs re-examining.
Removing customers with zero prior settled volume from both cohorts should shrink the observational estimate, since new-account fraud concentrates there.

Follow-up

Compliance and finance both object to deliberately approving transactions you believe are fraudulent. What is your answer, and how do you size the holdout?
Your interval spans the decision boundary. What do you recommend?
How would you detect that this number has gone stale?

Size a two-week test on the ecommerce authorization path

Easy · power · mde · ratio metrics · clustering

Your team wants to test a new decline-retry policy on the ecommerce channel. The metric is the dollar-weighted authorization approval rate from fct_payment_authorization: approved amount_minor over attempted amount_minor, after collapsing retry chains within a 15-minute window on (card_token_id, merchant_id, amount_minor) and excluding is_reversal rows and zero-amount verifications, all converted to one reporting currency. Baseline is 87 percent on roughly 900,000 collapsed attempts in two weeks. Randomisation is by customer_id, mean six attempts per customer. Give the minimum detectable effect at 80 percent power, two-sided alpha 0.05, and say what you would change if it is too large.

Approach

Compute the count-weighted binomial anchor first, because it is the number everyone expects and you need it to show why it is wrong. With n = 450,000 attempts per arm and p = 0.87, MDE = (1.96 + 0.8416) * sqrt(2p(1-p)/n).
State that the real metric is a ratio of two sums with a random denominator, not a Bernoulli mean, so the binomial standard error is not the right one. Linearise: Var(R) is approximately Var(Y_i - R * D_i) / (n * Dbar^2), where Y_i and D_i are the approved and attempted amount totals for customer i, n is the number of customers, and Dbar is the mean attempted amount per customer.
Apply the clustering correction, since assignment is by customer and attempts within a customer are correlated. Ask for or estimate the intra-customer correlation; the design effect is 1 + (m - 1) * rho with m = 6, so rho = 0.05 gives 1.25 and inflates the MDE by sqrt(1.25) = 1.118.
Do not assume the amount-weighted variance; estimate it by bootstrapping customers from the last eight weeks of history and resampling whole customers, which captures both the skew in amount_minor and the within-customer correlation in one step.
If the MDE exceeds the plausible effect, list the levers in order of cost: extend duration (MDE falls as 1/sqrt(n)), restrict to the segment where the rule can bind at all rather than diluting across all traffic, apply CUPED on the customer's pre-period approved amount, or switch the primary to the count-weighted rate and demote the dollar-weighted rate to a secondary read.

Follow-up

The rule only changes behaviour on declined attempts, which are 13 percent of traffic. How does restricting the analysis population to attempts that could have triggered the rule change both the MDE and the estimand?
How would your sizing change if the split were 90/10 instead of 50/50, and why is the loss more than proportional?
Currency: the metric sums amounts converted to one reporting currency. Would you use settlement_fx_rate or a rate table pinned at test start, and what does the choice do to variance?

Settled volume jumps after a new acceptance corridor launches

Medium · minor units · currency conversion · reconciliation

Settled payment volume per active customer rose 12 percent in the month a new acceptance corridor went live. The pipeline sums settlement_amount_minor from fct_payment_authorization and divides by 100 to get major units. Columns available: amount_minor, transaction_currency, captured_amount_minor, settlement_amount_minor, settlement_currency, settlement_fx_rate, settled_at, is_reversal, issuer_country, acquirer_country. Confirm or refute the 12 percent, and specify exactly the conversion the metric should be using.

Approach

Split the month-over-month increase by settlement_currency. If one currency carries nearly all of it while its transaction count barely moved, the problem is arithmetic rather than demand, and that is a two-minute check.
Look up each currency's ISO 4217 exponent. Dividing a three-decimal currency by 100 instead of 1000 overstates it tenfold; dividing a zero-decimal currency by 100 understates it hundredfold. Neither is a rounding issue.
Rewrite the conversion to scale settlement_amount_minor by ten to the power of that currency's exponent, then convert to the reporting currency. Do not reuse settlement_fx_rate for this; it is the rate applied between transaction and settlement currency at settlement time, not a reporting-currency rate.
Check the other half of the series too. settlement_amount_minor is denominated in settlement_currency while amount_minor is in transaction_currency, so any series that mixes the two is uninterpretable regardless of the exponent fix.
Exclude reversals and refunds, recompute, then reconcile the corrected total to the settlement ledger for the same period and show the tie-out in the write-up.
Recompute the per-customer denominator on the corrected month, since a new corridor adds customers as well as volume and the ratio can move either way once the numerator is right.

Follow-up

Where else in the warehouse does a hardcoded divide-by-100 appear, and how would you find every instance?
What test would fail the build the next time a currency with a different exponent is onboarded?
The reporting-currency rate is itself a choice. Which rate, on which date, and why does finance care about the 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.

Where your data came from

Data people depend on systems owned by other teams, and much of the job is negotiating for instrumentation, access, or a fix to a broken pipeline. Prepare an example of getting something changed upstream that you did not control. Describe what you asked for, what you traded, and how you worked while you waited.

Why do you want to apply your data science skills to the insurance sector, and specifically to AXA?

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

What would you do differently if you ran that project again?
How did you know the outcome was caused by your change?

Tell me about a time you had to explain a highly complex, "black-box" machine learning model to a non-technical stakeholder, such as an underwriter or actuary.

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.
Quantify the outcome, including what you would not claim credit for.

Follow-up

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

Retract a published number after finding a currency bug

Medium · error disclosure · minor units · process repair

Two weeks ago you published an interchange and fraud analysis that summed amount_minor across fct_payment_authorization without converting currencies. Minor units are not two decimals everywhere: some currencies carry none and some carry three, so the sum has no interpretation. A pricing decision is already in flight on the back of it. You now have corrected figures. Produce the retraction: what you send, to whom, in what order, and what you change in the process so this class of error is caught next time rather than trusted next time.

Approach

Size the error before announcing it, because saying the number is wrong without a magnitude and a direction forces every reader to assume the worst case.
Check whether the conclusion actually flips: if the ranking that drove the pricing decision is unchanged, that belongs in the first sentence beside the correction rather than buried at the end.
Tell the person acting on it first and directly, then the wider distribution, using the same text, so nobody learns about it secondhand.
Write the correction as four parts: the old number, the cause in one clause, the effect on the pending decision, and the new number. Leave out self-flagellation, which makes the reader do emotional work instead of acting.
Fix the class rather than the instance: a rule that a sum over amount_minor either groups by transaction_currency or passes through both conversion steps, exponent scaling and then a dated rate into one named reporting currency, plus a standing reconciliation of the settled subset to the settlement ledger inside each settlement_currency.

Follow-up

The corrected figures do not change the decision. Do you still send the correction, and what does that choice signal?
What automated check would have caught this, where would it live, and what would it cost in false alarms?