Infoorigin Data Scientist Interview Questions & Guide 2026

The 60-second brief

A Data Scientist at Infoorigin plays a pivotal role in transforming complex, unstructured datasets into high-impact business solutions. Operating at the intersection of advanced analytics, machine learning, and software engineering, you will design and deploy models that directly influence product roadmaps and client outcomes. The role is highly strategic, demanding a balance between classical statistical methods and cutting-edge artificial intelligence to solve real-world problems.

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

Focus areas

• Power experiments for heavy-tailed account revenue
• Separate contracted seats from actively used seats
• Read NRR on a fixed account cohort

Overview

At Infoorigin, data science is not an isolated academic exercise; it is deeply integrated into the core product lifecycle and client delivery pipelines. You will collaborate closely with cross-functional teams to build scalable data pipelines, design predictive models, and implement advanced technologies like Natural Language Processing (NLP) and Generative AI. Whether optimizing predictive maintenance algorithms or developing neural network architectures for complex text processing, your work will drive measurable efficiency and innovation.

To succeed in this position, you must possess strong technical foundations and a highly practical mindset. values data scientists who can look beyond theoretical accuracy and understand how models perform in production environments. Candidates who thrive here are those who take absolute ownership of their projects, from initial data preprocessing to final model evaluation and deployment.

The Infoorigin interview process

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

Written Test

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.

Technical Interview Rounds

This round decides whether someone can hand you a schema and a question and trust the number that comes back. Correctness under a clock is the bar, not clever syntax. The habit that separates strong from weak answers is checking the grain: after every join, know how many rows you expect and whether the count moved. Most wrong answers in this format are not wrong logic, they are a fan-out from a key that turned out not to be unique, or a filter applied before an aggregate when it belonged after. Say what you expect before you run it.

What to demonstrate

• Whether your row counts survive each join, and whether you notice on your own when they do not
• Deliberate handling of rows that fail to match, including whether the question needs an inner join or a left join with the non-matches kept and counted
• Whether NULLs are treated on purpose, given that a NULL compares equal to nothing and that COUNT of a column skips it
• Reaching a defensible answer inside the window instead of a refined one after it

How to prepare

• Take a two-table schema, write a join that fans out on purpose, then fix it by collapsing the many-side to one row per key before joining. Repeat until the fix is reflex rather than recall.
• Write a funnel as one query and print the distinct user count at each stage, then confirm each stage is a subset of the one above it rather than assuming it
• Do a few timed runs in a plain text box with no autocomplete and no formatter, since assessment editors often have neither

Behavioral Fit Assessment

Rounds of this kind usually include one question about work that did not go well, and it is the part that carries the most information. Anyone can narrate a shipped win. What the interviewer learns from a project that stalled is how you behave without a result to hide behind: whether you noticed the problem yourself, how long it took, and who you told. Answers that route the failure onto a data pipeline or a reorganisation close the topic without answering it, and the follow-up comes back to your own part.

What to demonstrate

• Whether you found the error yourself or someone else found it, and how long it sat before anyone knew
• What you changed afterwards, stated as a check you now run rather than a lesson you now believe
• Whether the mistake you choose has real cost attached, such as a quarter of misdirected roadmap or a metric that was reported upward, instead of one that flatters you

How to prepare

• Choose a failure you caught yourself and be ready to say what tipped you off. A story where someone else caught it is still usable, but you will be asked why you missed it.
• Write down the check you added afterwards and where it lives now, so the correction is a concrete artefact rather than a resolution.
• Rehearse saying the cost out loud. Candidates shrink the number by instinct once the interviewer is in the room.

Where candidates lose points

Randomising an experiment at the user level when users share an account

Two problems fire at once. Colleagues in one workspace see each other's work and talk to each other, so a treated user changes the behaviour of a control user in the same account, which violates the no-interference assumption and biases the estimate toward zero. Separately, outcomes within an account are strongly correlated, so the effective sample size is roughly n / (1 + (m - 1) * rho) for m users per account and intra-class correlation rho, not n. With rho around 0.3 and twenty users per account that is a design effect near 6.7, meaning a user-level confidence interval is about two and a half times narrower than it should be and results cross significance thresholds on noise alone. Randomise the account and cluster the standard errors.

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

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

Sizing estimates built on unnamed, unrevisable assumptions

Write each assumption as a named number you can change, then show the arithmetic so the interviewer can challenge one input instead of the whole answer. Finish by saying which assumption the result is most sensitive to, which matters more than the point estimate.

Crediting a treatment for regression to the mean

Selecting a group because it is extreme (lowest-engagement users, accounts having their worst month, the bottom decile of a score) moves that group's expected next-period value back toward the average even under no treatment, by exactly as much as the selecting measure is imperfectly correlated with its own later value. Compare against units that met the same selection rule and went untreated, or use two pre-periods so the bounce-back is visible before the intervention starts. A pre-post number on a group chosen for being extreme measures the selection rule, not the treatment.

Questions & practice

Explain the difference between L1 (Lasso) and L2 (Ridge) regularization. How do they physically impact the model's feature weights?

Medium · machine learning and modelling

Approach

1. Set a baseline first, so any model has something honest to beat.
2. Check what information would not exist at prediction time, and exclude it.
3. Pick an evaluation metric that matches the cost of each error type, not a default.

Follow-up

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

What is the vanishing gradient problem in deep neural networks, and how do activation functions like ReLU or architectures like LSTMs address it?

Medium · machine learning and modelling

Approach

1. Pick an evaluation metric that matches the cost of each error type, not a default.
2. Set a baseline first, so any model has something honest to beat.
3. Check what information would not exist at prediction time, and exclude it.

Follow-up

• What would you monitor after launch to know the model is still valid?
• Where could label leakage enter this setup?

How would you fine-tune a pre-trained Generative AI model for a specific domain-specific question-answering task?

Medium · machine learning and modelling

Approach

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

Follow-up

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

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

1. 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.
2. 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.
3. 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.
4. 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.
5. 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)

1. unchainable = df.idempotency_key.isna(); report df.loc[unchainable].groupby('account_id').billable_units.sum() before proceeding.
2. keys = ['account_id','endpoint','idempotency_key']; c = df[~unchainable].sort_values(keys + ['request_at'], kind='mergesort'); c['is5'] = (c.http_status >= 500).astype(int)
3. c['attempt_no'] = c.groupby(keys, sort=False).cumcount(); c['prior_5xx'] = (c.groupby(keys, sort=False).is5.cumsum() - c.is5) > 0
4. 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 difference between lists and tuples in Python, and describe scenarios where you would choose one over the other for memory optimization.

Medium · sql and data manipulation

Approach

1. State the window function and its partition and ordering out loud before writing it.
2. Compute rates by summing numerator and denominator separately, never by averaging rates.
3. Say which table is the grain you start from, and join outward from 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?

Write a Python function to identify and handle missing values in a dataset without using external libraries.

Medium · sql and data manipulation

Approach

1. Handle the rows that do not match: a LEFT JOIN with a NULL check is usually the question.
2. Compute rates by summing numerator and denominator separately, never by averaging rates.
3. 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?

How do generators work in Python, and how do they assist in handling large-scale datasets that exceed system memory?

Medium · sql and data manipulation

Approach

1. State the window function and its partition and ordering out loud before writing it.
2. Compute rates by summing numerator and denominator separately, never by averaging rates.
3. Say which table is the grain you start from, and join outward from it.

Follow-up

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

Monthly account margin joined as-of the live contract version

Medium · as-of join · fan-out · scd2 · unit economics

fct_usage_daily carries account_id, usage_date, net_amount_cents and cogs_cents. fct_subscription_period carries account_id, plan_tier, term_start_date, term_end_date, booked_at and is_current, with one row per contract version, so several superseded versions can bracket the same usage_date. dim_account is a type 2 dimension keyed on account_id with employee_band, is_internal, effective_from, effective_to and is_current. Return monthly net revenue, allocated COGS and gross margin per account, labelled with the plan_tier in force during that month and the employee_band in force at month end. The reported figure excludes internal accounts, so it cannot equal the raw sum of net_amount_cents; reconcile additively instead, quantifying every bucket you drop so the reported and dropped amounts add back to that raw sum.

Approach

1. Aggregate fct_usage_daily to one row per (account_id, month) before touching any dimension. Aggregating after the join is what multiplies revenue, and no amount of DISTINCT afterwards recovers the right number.
2. Resolve the contract as-of the month rather than as-of now. Take fct_subscription_period rows whose term brackets the month, then rank with ROW_NUMBER() OVER (PARTITION BY account_id, month ORDER BY booked_at DESC, subscription_period_id DESC) and keep rank 1. Filtering on is_current instead backdates today's plan over last year's usage and quietly rewrites history.
3. Resolve dim_account with the half-open interval effective_from <= month_end AND (effective_to > month_end OR effective_to IS NULL). The NULL on the live version has to be spelled out or the current row drops out of every recent month.
4. Compute margin as (sum(net_amount_cents) - sum(cogs_cents)) / NULLIF(sum(net_amount_cents), 0) at the account-month grain, and report the distribution rather than a blended rate so margin-negative accounts stay visible.
5. State the month-straddling rule explicitly: either split the month at the amendment date or take the version in force at month end. Both are defensible; an unstated choice is not.
6. Reconcile additively, not by equality to the unfiltered total. The output deliberately drops two populations: account-months whose as-of dim_account version carries is_internal = true, and account-months with no fct_subscription_period version bracketing the month, which an inner join on the contract removes without saying so. The statement that ties out is reported_net + internal_net + unmatched_net = sum(net_amount_cents) over the same usage_date range. Plain equality to the raw sum would only hold if both dropped buckets were empty, and the internal one never is.
7. Compute the internal bucket with the identical as-of rule used for the output, evaluating is_internal on the dim_account version in force at that month end. Reading is_internal from the current version instead moves accounts between the two sides of the identity and it stops closing.

Worked solution (30 min)

1. Build usage_monthly: SELECT account_id, date_trunc('month', usage_date) AS month, sum(net_amount_cents) AS net_cents, sum(cogs_cents) AS cogs_cents FROM fct_usage_daily GROUP BY 1, 2.
2. Build contract_asof by joining usage_monthly to fct_subscription_period on account_id with the term bracketing the month, then applying the ROW_NUMBER ranking on booked_at DESC and keeping rn = 1.
3. Join dim_account with the half-open effective_from/effective_to predicate evaluated at month end, and filter is_internal = false on the version selected.
4. Select account_id, month, plan_tier, employee_band, net_cents, cogs_cents and (net_cents - cogs_cents)::numeric / NULLIF(net_cents, 0) AS gross_margin.
5. Run the reconciliation as a three-way identity over the same usage_date range: sum(net_cents) in the output, plus sum(net_cents) over account-months whose as-of dim_account version has is_internal = true, plus sum(net_cents) over account-months with no bracketing subscription version, must equal SELECT sum(net_amount_cents) FROM fct_usage_daily for that range, to the cent. Publish the two dropped amounts next to the total rather than leaving them implicit; a non-zero unmatched bucket is a contract-coverage bug to chase, not rounding.

Expected result: One row per account per month with net revenue, allocated COGS, gross margin, and the plan_tier and employee_band that were actually in force, with output row count equal to the number of distinct (account_id, month) pairs that survive both the internal-account filter and the contract-bracket join.

Verify it yourself

• Output row count equals COUNT(DISTINCT (account_id, date_trunc('month', usage_date))) on the source fact restricted to the same two filters; counted against the unrestricted fact it is strictly smaller, by exactly the internal and unmatched account-months.
• reported_net + internal_net + unmatched_net equals the raw sum from fct_usage_daily over the same usage_date range, to the cent, with all three terms computed from the same usage_monthly CTE.
• The reported figure alone is strictly below the raw sum whenever any internal account recorded usage in the window, and equal to it only in a window where none did.
• No account-month carries more than one plan_tier or employee_band value.
• Accounts with net_cents of zero return NULL margin rather than a division error, and margin-negative accounts appear with a value below zero rather than being dropped.

Follow-up

• An account amends mid-month from team to enterprise. Show what your query reports and argue for one attribution rule over the other.
• Your monthly total disagrees with the finance figure by a small amount. Where would you look first, and which number do you defend?

Triage a sample ratio mismatch before reading the result

Easy · srm · assignment · instrumentation

An account-randomised test assigned arms 50/50 by hashing account_id. Exposure logging recorded 4,200 accounts: 1,987 in treatment and 2,213 in control. A chi-square goodness-of-fit test against the intended 2,100/2,100 split gives chi-square(1) = 12.2, p about 0.0005. The analyst reports the primary metric up 6% in treatment and asks to ship. Explain what the imbalance implies about that 6%, rank the causes you would investigate first given this domain's data, and name for each the single query against fct_api_request or dim_account that confirms or eliminates it.

Approach

1. State the rule before diagnosing anything: a sample ratio mismatch below a 0.001 alarm threshold invalidates the readout. Whatever mechanism removed 113 accounts from one arm almost certainly removed them non-randomly, which makes it a selection effect on the outcome. The 6% is not to be adjusted, caveated or shipped; it is unusable until the mechanism is named.
2. Verify the test is testing the right thing. chi-square = (1987 - 2100)^2 / 2100 + (2213 - 2100)^2 / 2100 = 6.08 + 6.08 = 12.16 on one degree of freedom. Confirm the denominator is the intended assignment count rather than the observed total, and that you are testing assignment rather than an analysis population that has already been filtered.
3. Rank causes by how this domain actually breaks rather than by textbook order: is_internal accounts filtered after assignment instead of before; exposure logged on a code path that returns 5xx more often in one arm, silently dropping those accounts; assignment recorded at first request but exposure at a later event, so accounts that churned in between appear in only one arm; dim_account type-2 versioning giving one account_id two surrogate keys and two hash inputs; deployment_model = 'self_hosted' accounts that never emit exposure telemetry at all.
4. Attach one discriminating query to each: count arms before the is_internal filter; compare the http_status >= 500 rate by arm on the exposure endpoint in fct_api_request; compare the distribution of created_at and churned_at by arm in dim_account; count distinct account_sk per account_id inside the experiment window; cross-tabulate arm against deployment_model.
5. Compare counts at three fixed stages, assignment, first exposure, and analysis population, and localise the divergence to one of them. The stage where the arms first separate names the subsystem, and everything downstream of it is a symptom.
6. Report the outcome as abort, fix, rerun, and be explicit about what survives: the variance estimate for re-powering, the instrumentation fix, and nothing whatsoever about the effect size.

Follow-up

• The imbalance disappears once you restrict to accounts with at least one successful request. Does that fix the experiment or confirm the bug?
• What alarm threshold would you set for this check, and why is 0.05 the wrong one for a diagnostic you run on every experiment every day?
• How would you detect a mismatch confined to one segment while the overall split looks clean?

Measure switching cost when switching cost is not observable

Medium · proxy metrics · bias statement · integration depth · stratification

Leadership wants switching cost tracked as a leading indicator of renewal. Nothing in the warehouse records it. Available: fct_api_request (account_id, workspace_id, environment, api_key_id, sdk_name, sdk_version, endpoint, traffic_class, http_status, request_at) and dim_account (account_id, employee_band, deployment_model, is_internal). Propose a proxy, state the direction and likely size of its bias, and name one decision the proxy is good enough for and one it is not. Deliver the proxy definition, the written bias statement, and the validation you would run against observed renewal outcomes.

Approach

1. Say first that switching cost is unobservable in this data and that the deliverable is a biased correlate with its bias written down, not a measurement. Anything presented as a direct measure of switching cost is already wrong before the SQL starts.
2. Define the proxy as production integration breadth per account over a trailing 28 days: distinct endpoint route templates, distinct api_key_id, and distinct workspace_id with environment = 'production', all restricted to http_status < 400 and traffic_class in ('interactive','batch'). Breadth, not volume, because request volume is one CI configuration change away from an order of magnitude.
3. State the biases with their mechanisms and their sign. Upward with account size, because breadth correlates with employee_band, so the proxy ranks large accounts as sticky whether or not they depend on anything. Downward for deployment_model = 'self_hosted', whose traffic does not all cross the managed gateway, so their breadth is systematically understated. Blind to criticality: one endpoint carrying a production billing path is a larger switching cost than twenty endpoints behind a read-only dashboard, and nothing in this data distinguishes them.
4. Handle the bias by stratifying rather than by pretending it is gone. Report the proxy within employee_band and deployment_model strata and state explicitly that cross-stratum comparisons are not supported by the construction.
5. Validate against the only outcome that is actually observed: renewal on the renewal-eligible base. Within strata, report renewal rate by proxy quintile for accounts whose term_end_date has passed plus a 45-day grace, and report discrimination at the operating point a capacity-bound team can work rather than a global AUC.

Worked solution (30 min)

1. Compute the trailing-28-day proxy per account: distinct endpoint, distinct api_key_id and distinct production workspace_id under successful, interactive-or-batch filters.
2. Join dim_account for employee_band and deployment_model and drop is_internal = true.
3. Build the renewal-eligible cohort from fct_subscription_period term_end_date in the target months with a 45-day grace, labelling each account renewed or not.
4. Within each employee_band stratum, report renewal rate by proxy quintile, and report the proxy distribution for self_hosted accounts separately so its understatement is visible.

Expected result: A proxy defined as production integration breadth over a fixed 28 days, a written bias statement naming upward bias with account size, downward bias for self-hosted deployment and total blindness to criticality, plus a within-stratum validation showing renewal rate by proxy quintile. The answer names one supported decision, such as ordering a capacity-bound outreach list inside a stratum, and one unsupported decision, such as setting price or contract terms by switching cost.

Verify it yourself

• Every account's proxy must be computed over the same 28-day window; a variable window makes the quintiles incomparable.
• Self-hosted accounts should show systematically lower breadth; if they do not, gateway coverage of that deployment model differs from what was assumed and the bias statement is wrong.
• The renewal-eligible cohort should be roughly one twelfth of the paying base per month on annual contracts; a much larger share means the term_end_date filter is not doing what you think.

Follow-up

• An account's endpoint breadth drops 40% in a week. Name three explanations that have nothing to do with reduced dependency.
• Which decision would you refuse to make on this proxy, and what evidence would you need before making it?

Net revenue retention jumps sixteen points in one month

Medium · versioned tables · fan-out · metric definition

Trailing-twelve-month net revenue retention printed around 108 percent for months and now reads 124 percent, with no unusual deals closed. The query sums arr_cents from fct_subscription_period (account_id, arr_cents, term_start_date, term_end_date, amendment_type, superseded_by_id, is_current, booked_at) filtered on is_current = true at month M, across accounts holding arr_cents > 0 at month M-12. Find the defect, correct the number, and rewrite the definition so the next person cannot reintroduce it.

Approach

1. Audit the grain before the arithmetic: count account_ids holding more than one row with is_current = true and superseded_by_id null. A versioned contract table that double counts one amendment batch inflates the numerator while leaving the denominator untouched.
2. Replace is_current with an as-of selection on both dates, taking the version whose term_start_date and term_end_date bracket the reporting date and tie-breaking on latest booked_at. The numerator is read as of M and the denominator as of M-12; neither uses today's live version.
3. Verify the cohort is frozen. The account set is fixed at M-12 and nothing acquired since may enter the numerator, so check that no join to a current-period table quietly re-admits new accounts.
4. Confirm the estimand is a ratio of sums rather than a mean of per-account ratios. Contraction is floored at zero while expansion is unbounded, so the two constructions differ systematically and the second is far noisier.
5. Reissue the definition with the failure modes written into it: exactly one row per account per date by construction, cohort frozen at M-12, churned accounts contributing zero rather than dropping out of the numerator.

Follow-up

• A churned account should contribute zero rather than disappear. What does the ratio do under each treatment, and which one is correct?
• How would you unit-test this metric so a future amendment batch with the same defect fails a check instead of reaching a board slide?

Suggested practice checklist

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

Day 1 — Design one test end to end on paper

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

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

Day 2 — Power arithmetic until it is automatic

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

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

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

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

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

Day 4 — Validity threats you can actually test for

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

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

Day 5 — When randomization is not available

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

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

Day 6 — The readout query

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

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

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

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

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

Being wrong in public

A number you shipped turned out to be wrong, and someone had already acted on it. That is one of the most useful stories a data person can carry. What is being scored is how fast you noticed, who you told first, and what you changed in the process so the same class of error could not repeat quietly.

Walk through your typical data preprocessing pipeline. How do you handle highly skewed numerical features and high-cardinality categorical variables?

Medium · behavioural and stakeholder questions

Approach

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

Follow-up

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

Disagree with a product manager about an adoption claim

Medium · conflict · telemetry hygiene · adoption metrics

A product manager is about to present that a new SDK release drove a 40 percent rise in requests among adopting accounts, computed from fct_api_request counts grouped by sdk_version. You find the rise is concentrated in traffic_class equal to ci, that rows with is_retry true grew alongside it, and that restricting to interactive non-retry traffic leaves a 3 percent lift. The launch review is in two days. Decide how you raise this, with whom and in what order, and what you propose the claim becomes.

Approach

1. The interviewer is probing whether you can correct a colleague without ambushing them, and whether your own counter-analysis carries the caveats theirs lacked. Go to the product manager privately before the review. A correction delivered in the room is a status move and loses the argument you are actually trying to win.
2. Bring a decomposition rather than a verdict: the same accounts and window, requests split by traffic_class with retries held out as their own column, so their 40 percent and your 3 percent reconcile line by line and neither has to be taken on trust.
3. Reproduce their figure exactly first. If you cannot land on 40 percent with their method, you do not yet know what you are disagreeing with.
4. Ask whether the continuous-integration lift is itself valuable. An account wiring the SDK into its pipeline has increased integration depth, which is the dominant switching cost in this domain, so the honest claim may be that integration depth rose while interactive usage moved 3 percent. Improving the claim beats deleting it.
5. Name the mechanism that makes the raw count dangerous: clients retry when the platform degrades, so retry volume climbs exactly when the customer is most at risk. Pull the 5xx rate for the same accounts and window before anyone concludes anything, and note that billable_units is zero on 5xx rows, so request counts and billable quantities diverging is itself the signal.
6. Close with a standing definition for launch metrics so the next release does not repeat the exercise.

Follow-up

• The product manager argues that continuous-integration traffic is real usage and declines to split it out. Is that position defensible, and under what metric definition?
• Suppose the 5xx rate for those same accounts also rose 40 percent. What is the claim now?
• The review happens and the raw number is presented regardless. What do you do next, and what do you not do?

Walk through an analysis you later discovered was wrong

Easy · data quality · error ownership · metering

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

Approach

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

Follow-up

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