The role of a Data Scientist at UC San Francisco is pivotal in leveraging data to drive decisions that impact healthcare outcomes, research advancements, and operational efficiencies. As a leader in biomedical research and patient care, UCSF utilizes data science to analyze complex datasets, develop predictive models, and derive actionable insights that enhance patient care and inform public health policies. You will work alongside interdisciplinary teams, contributing to projects that may involve genomics, epidemiology, or clinical trials, making your work both impactful and meaningful.
In this role, you will engage with cutting-edge technologies and methodologies to tackle challenging problems that affect real lives. The scale and complexity of the data you will handle, combined with UCSF's commitment to innovation, will provide you with a stimulating professional environment. Expect to influence key initiatives that shape strategies across various departments, driving the mission of UCSF to improve health and healthcare through data-driven insights.
Phone Screen
reportedA screening call is a matching exercise run by someone who will not evaluate your statistics. They are checking that the work described on your resume is work you personally did, and that its scope matches the level the role is written for. Logistics get settled in the same half hour so nobody spends an interviewer's afternoon on a mismatch. The answer that fails is the one narrated in the plural. If every sentence is 'we built' and 'the team decided', there is nothing specific to write down about you. Name the piece that was yours, the decision you made inside it, and what changed after.
What to demonstrate
- Whether the ownership implied by your resume survives one round of follow-up about who actually did which part
- Whether your described scope (data size, stakeholders, what shipped) matches the seniority the role is written at
- Whether timeline, location and compensation expectations make the rest of the loop worth scheduling
How to prepare
- Rewrite your top three resume bullets in the first person singular, each with the decision you made and what moved afterwards, then say them out loud once so the 'we' does not return under pressure
- Attach one number to each project: the baseline, the change, and the window it was measured over. Where impact was never measured, say that plainly rather than inventing a figure
- Settle your compensation range before the call and give it as a range with a reason behind it, such as current total comp or a competing timeline, instead of deflecting the question twice
Technical Interview
reportedBefore anything else, this round is a reading test. You are given a small schema and a question phrased in business language, and most of the difficulty sits in the gap between them. Who counts as an active user, does a refunded order still count as an order, is that date column an event time or a load time. Weak answers start typing immediately and compute something precise about the wrong population. Strong ones pin the definition in one sentence, name the column that encodes it, then write the query. On a timed assessment with nobody to tell, write the definition in a comment anyway.
What to demonstrate
- Whether an ambiguous term becomes a specific column and filter before any computation happens
- Whether you read the schema for keys and cardinality rather than only for column names
- Whether the result answers the question at the grain it was asked at, per user or per session or per day
How to prepare
- Take three metrics you already use and write down the exact filter and exact grain behind each, then practise stating one of them in a single sentence out loud
- On a schema you have never seen, spend the first minute writing what one row of each table means and which key it is unique on, then predict which joins can duplicate rows
- Rehearse a version where the definition changes halfway through, and edit the query you have instead of starting over
Onsite Interview
reportedA day of back-to-back interviews samples your floor, not your ceiling. Four hours in, the habits that carry a good answer are the first to go: restating the question before solving it, asking what the data would have to look like, checking a number before quoting it. What the day decides is whether the tired version of you is still someone to leave alone with an ambiguous problem. The round that sinks a candidate is usually not the hardest one. It is the one immediately after the round that went badly.
What to demonstrate
- Whether the late rounds get the same clarifying questions as the first one, or whether you start answering immediately to save effort
- Whether a weak answer stays in the room it happened in, instead of following you into the next conversation as apology or distraction
- Whether the quality of your questions holds up, since fatigue removes curiosity about the problem before it removes knowledge of the method
How to prepare
- Rehearse the length, not just the content: book four mock interviews of different types in one afternoon with short gaps, because the one you need to observe is the fourth
- Put the two or three questions you ask at the start of any problem on a card in front of you, so that under fatigue it is a habit you run rather than a decision you make
- Decide in advance what the gap between rooms is for: water, one line of notes on anything you promised to follow up, and an explicit close on the round that just ended so it does not travel
- Prepare a different closing question for each interviewer, so the end of a long day does not produce the same one four times
PracHub editorial advice for the preparation topics above.
Treating clinical measurements as missing at random
A lab result, a vital sign, or a screening exists because someone ordered it, and ordering tracks suspicion of disease, visit frequency, and site workflow. Imputing the mean or dropping incomplete rows biases the population estimate and can flip the sign of an association, because the untested are systematically healthier or systematically disengaged. The presence indicator is often more predictive than the value, which is a warning sign rather than a feature win: a model that learns test ordering will not transfer to a site with different protocols.
Rates built on member counts rather than exposure
Members join and leave mid-period, so dividing events by distinct members mixes a person covered for 30 days with one covered for 365. New joiners also have artificially low observed utilisation because their claims have not arrived yet and because care takes time to initiate. Denominators must be member-months or member-years, and comparative quality measures usually need a continuous-enrolment requirement with an explicit allowable gap, stated in days.
Writing SQL without stating NULL and tie-breaking behaviour
Before calling a query finished, say what it does with NULLs, ties and empty groups. NOT IN against a subquery containing a single NULL returns no rows at all, and RANK, DENSE_RANK and ROW_NUMBER differ precisely on ties, so name which one the question requires.
Ending an analysis without a recommendation or next step
Close with what you would do and what would change your mind, stated as a condition you can check later. If the evidence is genuinely inconclusive, recommend the specific next measurement and say what it costs in time or exposure.
Choose a category, try a prompt, then open its approach, worked solution or follow-up when you need it.
What metrics do you use to evaluate the performance of a machine learn…
What metrics do you use to evaluate the performance of a machine learning model?
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
- 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?
Create a simple predictive model for a given dataset and explain your …
Create a simple predictive model for a given dataset and explain your reasoning.
Approach
- Set a baseline first, so any model has something honest to beat.
- Frame the prediction: the label, the moment of prediction, and the action it triggers.
- Say how the offline result would be validated online before it is trusted.
Follow-up
- Where could label leakage enter this setup?
- What would you monitor after launch to know the model is still valid?
Simulate how ignoring clustering inflates the false positive rate
Simulate a provider-level comparison with no true effect. 40 providers, 60 members each, a continuous outcome with total variance 1 and intraclass correlation 0.02. Randomise providers 20 to 20, not members. Over 5,000 replications, report the share of replications where a two-sample t-test run on all 2,400 individual observations, ignoring provider, returns p below 0.05. Report the same share when the unit of analysis is the 40 provider means. Then state how the first number follows from the design effect 1 + (m - 1) x ICC.
Approach
- Decompose the variance explicitly rather than tuning it: provider variance = ICC = 0.02, residual variance = 1 - ICC = 0.98. Draw u_j once per provider and e_ij per member, so the outcome is u_j + e_ij and the correlation between two members of one provider is 0.02 by construction.
- Randomise at the provider level. That is the entire mechanism: the treatment indicator is constant within a provider, so in any single replicate the provider intercepts are confounded with the arm, and the naive test reads that confounding as signal.
- Vectorise over replications with a (reps, providers, members) array. 5,000 x 2,400 normals is trivial in numpy, and a Python loop is what tempts people to cut replications to 500 and report a number with Monte Carlo noise of plus or minus 0.017.
- Predict the answer before running it: design effect = 1 + (60 - 1) x 0.02 = 2.18, naive standard errors are too small by sqrt(2.18) = 1.476, so the rejection rate at nominal 0.05 is about 2 x (1 - Phi(1.96 / 1.476)) = 0.18. Agreement between prediction and simulation is what makes the result a demonstration rather than an anecdote.
- Run the provider-mean analysis as the control. Recovering 0.05 there proves the generator is correct and isolates the failure to the analysis unit.
Worked solution 30 min
- Draw u with shape (reps, 40) from N(0, sqrt(0.02)) and e with shape (reps, 40, 60) from N(0, sqrt(0.98)); y = u[:, :, None] + e.
- Assign providers 0-19 to one arm and 20-39 to the other; because there is no true effect, a fixed split is valid and removes one source of Monte Carlo noise.
- Naive test: flatten each arm to 1,200 observations and compute a two-sample t statistic per replication with vectorised means and pooled variance.
- Cluster test: average within provider to 40 values, then run the same two-sample t on 20 versus 20.
- Report both rejection shares and compare the naive one against 2 x (1 - Phi(1.96 / sqrt(2.18))).
Follow-up
- How many providers would you need for 80 percent power on a 0.2 SD difference with 60 members each and this ICC?
- Cluster sizes are unequal in reality. The design effect approximation becomes 1 + ((1 + CV^2) x mbar - 1) x ICC. Which direction does that move your sample size and why?
- The intervention cannot be withheld from any provider. Sketch a design that still yields a defensible estimate.
Write a function to find the median of a list of numbers.
Write a function to find the median of a list of numbers.
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.
- Compute rates by summing numerator and denominator separately, never by averaging rates.
Follow-up
- How does the query change if the join becomes one-to-many?
- What breaks if events arrive late or out of order?
Count thirty-day unplanned readmissions without counting transfers
encounter holds encounter_id, patient_id, facility_id, encounter_type, admit_ts, discharge_ts, admission_type, discharge_disposition, is_planned_admission and index_encounter_id. For inpatient stays discharged between 2025-01-01 and 2025-11-30, return the 30-day unplanned readmission rate overall and by facility_id. Exclude an index stay when is_planned_admission is TRUE or discharge_disposition is transfer_acute, expired, hospice or ama. A qualifying readmission is any unplanned inpatient admission for the same patient with admit_ts after the index discharge_ts and within 30 days of it. One stay can be both a readmission and the next index stay.
Approach
- Decide between LEAD and EXISTS explicitly and say why. LEAD over the patient's admission stream gives the literal next stay, so an intervening planned admission hides a later unplanned one. The measure asks whether any unplanned admission fell in the window, which is an EXISTS or a self-join with a range predicate.
- Exclude transfers at the index end before anything else. An acute-to-acute transfer closes one encounter and opens another at a different facility_id within hours, so without the transfer_acute exclusion the facilities that move the sickest patients out look the worst.
- Apply the index exclusions as a set, including stays with NULL discharge_ts, which are still open and have no measurable window.
- Stop the denominator early enough that every index stay had a full 30 days of observation plus encounter lag. That is why the window ends 2025-11-30 rather than at year end.
- Compute the rate at stay level, not patient level, then compare the raw facility rates against an observed-over-expected framing before anyone publishes a ranking.
Worked solution 30 min
- CTE ip: inpatient encounters with a non-null discharge_ts.
- CTE index_stays: ip rows with discharge_ts in the window, is_planned_admission FALSE, and discharge_disposition not in the four excluded values.
- Flag each index stay with EXISTS against ip for the same patient_id where is_planned_admission is FALSE, encounter_id differs, admit_ts > index.discharge_ts and admit_ts <= index.discharge_ts + 30 days.
- Aggregate numerator and denominator overall and grouped by facility_id.
- Re-run with the transfer exclusion removed and record how much the rate moves.
Follow-up
- A patient has three stays ten days apart. How many index stays and how many readmissions, and why does the answer depend on the stated rule?
- Which comparison would you publish across facilities, raw rate or observed over expected, and what does risk adjustment still fail to fix?
- index_encounter_id already exists on the table. Would you trust it, and how would you audit it against your own logic?
You are given a dataset with various health metrics. How would you app…
You are given a dataset with various health metrics. How would you approach analyzing this data?
Approach
- Fix the population and the time window before naming any metric.
- State what result would change your recommendation, so the answer is falsifiable.
- Restate the decision this analysis has to support, and who acts on the answer.
Follow-up
- What would you do if the primary metric and the guardrail moved in opposite directions?
- How would you detect that the metric is being gamed rather than genuinely improving?
A healthcare provider is experiencing a drop in patient satisfaction. …
A healthcare provider is experiencing a drop in patient satisfaction. What data would you analyze to identify the issue?
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?
- What would you do if the primary metric and the guardrail moved in opposite directions?
How do you prioritize tasks when managing multiple projects?
How do you prioritize tasks when managing multiple projects?
Approach
- Restate the decision this analysis has to support, and who acts on the answer.
- Name one primary metric, then the guardrail that stops it being gamed.
- Decompose the metric into the rates that drive it, and say which one you would check first.
Follow-up
- What would you do if the primary metric and the guardrail moved in opposite directions?
- Which segment would you cut first, and what would that rule out?
Given a dataset, how would you optimize your code for efficiency?
Given a dataset, how would you optimize your code for efficiency?
Approach
- Restate the decision this analysis has to support, and who acts on the answer.
- Name one primary metric, then the guardrail that stops it being gamed.
- 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?
- What would you do if the primary metric and the guardrail moved in opposite directions?
How would you approach designing an experiment to test a new treatment…
How would you approach designing an experiment to test a new treatment's efficacy?
Approach
- Say whether units interfere with each other, and switch design if they do.
- State the primary metric and the minimum effect worth shipping, then size the test.
- Name the guardrails that would stop a launch even on a positive primary result.
Follow-up
- What would you conclude if the result is positive but the test is underpowered?
- What would you do if you could not randomise at all?
Describe a data science project you have worked on. What was your role…
Describe a data science project you have worked on. What was your role?
Approach
- Say what you would check first and why it is the highest-information step.
- State your assumptions explicitly before working the problem.
- Work from the decision backwards to the evidence you would need.
Follow-up
- How would you know your answer was wrong?
- What assumption would you test first?
Critique a denial rate slide before it is shown
A weekly slide says first-pass denial rate improved from 9.1 percent to 7.4 percent month over month. It was computed from medical_claim_line as rows with claim_status 'denied' divided by all rows with a received_date in the month. You cannot query anything. List every assumption the number rests on, restate the metric so two analysts computing it independently would get the same value, and give the three questions you would ask before the slide is presented.
Approach
- Work the grain first. The denominator counts all rows, so a claim resubmitted after a denial contributes its original line, its replacement (frequency_code 7) and possibly a void (8). Both numerator and denominator are inflated by exactly the claims most likely to have been denied, which makes the rate a function of resubmission behaviour rather than adjudication.
- Work maturity second. A line received late in the month may still be pended at extract time, and pended lines are neither paid nor denied. If pends resolve to denials later, the most recent month is understated by construction, which looks exactly like improvement. Ask for the paid-through date and the adjudication runout hold.
- Work the unit third. A denial rate on claim lines and a denial rate on claims are different numbers, and a single denied line on a twenty-line facility claim is a different operational event from a wholly denied professional claim. The spec has to pick one and say so.
- Work mix fourth. Denial reasons have different owners and different volumes: eligibility denials belong to enrollment operations, coding denials to provider relations. A shift in submission mix across providers, place of service or reason code moves the blended rate with no change in any component, so the restated metric reports by denial_reason_code group.
- Work precision last. A 1.7 point move is decisive on two million lines and meaningless on three thousand. The restated metric carries the denominator count, and a slide without it cannot be read.
Worked solution 15 min
- Write the assumption list as a numbered set: single version per claim, adjudication complete at extract, line grain intended, stable reason-code and provider mix, denominator large enough for the move to be real.
- Restate the metric with numerator as lines denied at first adjudication, denominator as lines received in the month counted at claim_version 1, reported by received_date month, held 60 days, broken out by denial_reason_code group, and published with the denominator count.
- Write the three questions in the order that would kill the slide fastest: what is the paid-through date, how many lines are in each month's denominator, and did the reason-code mix move.
- State how the restated number is presented, with the most recent month marked provisional and restated at each refresh.
Follow-up
- The rate is still down after version dedup and a 60-day hold. What is the next thing you check before anyone celebrates?
- Which of these problems would also affect a paid-amount series computed the same way?
- How would you present the restated number so the audience understands the prior month is provisional?
Pooled persistency fell but no one lost coverage
Pooled 12-month enrollment persistency fell from 0.78 to 0.66 for the cohort anchored in a single month. member_enrollment carries span_id, member_id, plan_id, product_type, effective_date, termination_date, termination_reason_code, dual_eligible_flag and region_code. The metric allows at most one gap of 45 days or fewer. Product mix in that anchor month was not stable. Determine how much of the fall is real disenrollment rather than span bookkeeping or mix, and deliver a persistency series a contract owner could act on.
Approach
- Inspect the query's unit of analysis first. Spans are not members, and a benefit or plan change closes one span and opens the next on the same day, so any logic that treats a non-null termination_date as loss of coverage counts an administrative event as churn.
- Rebuild member-level coverage by unioning all spans per member and merging adjacent or overlapping ones, then apply the 45-day gap rule to the merged intervals rather than per span.
- Cut the terminations by termination_reason_code. plan_change and eligibility_redetermination have entirely different meanings for a contract, and death is not churn you can address.
- Split by product_type and stop pooling. Commercial open enrollment and managed medicaid redetermination produce structurally different churn on different calendars, and the pooled average of the two is a number that describes nobody.
- Compare the anchor month to the same month in the prior year rather than to the adjacent month, because enrollment is strongly seasonal and an anchor month sitting on an open-enrollment or redetermination boundary is not comparable to its neighbour.
- Apply the Kitagawa decomposition across product_type to separate how much of the 12 point fall is mix, meaning the anchor cohort simply contained more of a high-churn product, from how much is within-product deterioration.
Follow-up
- Why 45 days, and what would change if the allowable gap were 30 or 60?
- A member terminates and re-enrolls 90 days later under a different plan_id. Are they retained? Defend your answer against the opposite one.
- What guardrail would catch someone improving persistency by declining to enroll members likely to churn?
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.
Prepare, practise & reflect
One practical outcome each day. Spend longer where you need it.
0 / 7 done01Design 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.
Practice prompt ↗Practice prompt ↗Practice prompt ↗Worked solution ↗02Power 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.
Practice prompt ↗Practice prompt ↗Practice prompt ↗03Variance 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.
Practice prompt ↗Practice prompt ↗04Validity 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.
Practice prompt ↗Practice prompt ↗Worked solution ↗05When 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.
Practice prompt ↗Practice prompt ↗06The 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.
Practice prompt ↗Practice prompt ↗07Present 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.
Practice prompt ↗Practice prompt ↗Worked solution ↗Expand any day for tasks and deliverables. Your progress is saved on this device.
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.
How do you handle feedback and criticism?
How do you handle feedback and criticism?
Approach
- Pick a story where you drove the decision, not one where you observed it.
- Close with what you would do differently, concretely.
- Quantify the outcome, including what you would not claim credit for.
Follow-up
- How did you know the outcome was caused by your change?
- What would you do differently if you ran that project again?
Defend a null result against a programme sponsor
A care management programme enrolled the top 1 percent of members by prior-year allowed spend. The sponsor's deck shows allowed PMPM for enrollees falling 34 percent from the year before enrollment to the year after, and asks for budget to triple the programme. You rebuild the evaluation with a concurrent comparison group selected by the identical spend rule in the same period. The difference-in-differences estimate is a 3 percent reduction with a confidence interval spanning zero. Deliverable: how you present this, to whom, in what order, and what you propose next.
Approach
- Name the probe: whether you can deliver a finding that costs somebody their programme without softening it into uselessness or creating an adversary who routes around you next time.
- Lead with the mechanism, not the verdict. Show the comparison group's own unadjusted drop, which will be large, because a cohort selected on an extreme of the outcome regresses toward the mean whether or not anyone intervenes. The sponsor's 34 percent is mostly that, and it is a property of the selection rule rather than a criticism of their clinicians.
- Give the sponsor the finding privately before it appears in any deck their leadership sees. Being surprised in a room is what turns a methods disagreement into a political one.
- State the estimate with its interval and say what it rules out as well as what it fails to establish. A 3 percent point estimate whose interval crosses zero is not evidence of no effect; it is insufficient power to separate a modest effect from none, and those are different claims.
- Arrive with a design rather than only an objection. Propose a regression discontinuity at the enrollment threshold if the rule is applied sharply, or a randomised rollout across the next wave of eligible members, and state the sample size needed to detect the effect size the sponsor believes in.
Follow-up
- The sponsor says withholding the programme from a comparison group is unethical. What do you propose instead?
- Leadership wants one number for the board next week. What do you give them?
- What result would change your mind and make you believe the 34 percent?
Push back on an over-broad extract without blocking the work
A researcher asks for a one-off extract covering 12,000 members: member_id, date of birth, five-digit residential postal code, full lab_result rows including value_text, and clinical note text, held on their laptop for a retrospective analysis of one lab analyte over time. They describe it as de-identified because names are excluded. A flat refusal is not an acceptable answer. Deliverable: the column list you will actually provide with a justification per retained field, the sentence explaining why free text defeats an enumerated-identifier approach, your small-cell rule for the published stratification, and what you deliver this week.
Approach
- The probe is whether privacy is a design skill for you or a compliance reflex. Flat refusal and unquestioned fulfilment both fail, for opposite reasons.
- Reconstruct the analysis from the research question rather than from the request. A longitudinal trend in one analyte needs a stable pseudonymous subject key, the analyte's loinc_code, value_numeric, units, the reference bounds, and a date. It does not need names, notes, or a precise birth date.
- Replace each identifier with the least precise version that still answers the question: a salted per-study pseudonym instead of member_id, age in years or an age band instead of date of birth, a consistent per-subject date offset if only intervals matter, and a coarser geography than five-digit postal code.
- Explain the free-text problem in one sentence. Removing an enumerated list of identifier fields does not de-identify a note, because notes carry names, relationships, employers, dates and rare clinical detail in prose, so only a reviewed extraction into structured fields or a formal statistical determination addresses it.
- Set the publication rule before the analysis runs. Suppress cells below a stated threshold and suppress complements too, because a reader holding a row total and all but one cell recovers the suppressed count by subtraction.
- Keep them moving. Deliver the reduced extract into a controlled environment this week rather than onto a laptop, and offer to derive any note-based variable yourself as a structured flag so they never need the prose.
Follow-up
- They say the age band loses resolution their model needs. What do you do?
- How do you set the suppression threshold, and who signs off on it?
- The analysis later needs linking to a second data source. What changes?
- 01
How do you handle feedback and criticism?
- 02
A care management programme enrolled the top 1 percent of members by prior-year allowed spend. The sponsor's deck shows allowed PMPM for enrollees falling 34 percent from the year before enrollment to the year after, and asks for budget to triple the programme. You rebuild the evaluation with a concurrent comparison group selected by the identical spend rule in the same period. The difference-in-differences estimate is a 3 percent reduction with a confidence interval spanning zero. Deliverable: how you present this, to whom, in what order, and what you propose next.
- 03
A researcher asks for a one-off extract covering 12,000 members: member_id, date of birth, five-digit residential postal code, full lab_result rows including value_text, and clinical note text, held on their laptop for a retrospective analysis of one lab analyte over time. They describe it as de-identified because names are excluded. A flat refusal is not an acceptable answer. Deliverable: the column list you will actually provide with a justification per retained field, the sentence explaining why free text defeats an enumerated-identifier approach, your small-cell rule for the published stratification, and what you deliver this week.
Is this an official UC San Francisco interview guide?
No. It is PracHub's own research and practice material for the Data Scientist role at UC San Francisco. Rounds and questions reflect what candidates have reported, not a process UC San Francisco has published, and they change over time. Confirm the current format and scope with your recruiter.
PracHub interview research ↗How difficult are the interviews, and how much preparation time is typical?
The interviews can vary in difficulty, with technical assessments being more challenging. Candidates typically spend several weeks preparing, focusing on both technical skills and behavioral questions.
PracHub interview research ↗What differentiates successful candidates?
Successful candidates often demonstrate strong technical proficiency, effective communication skills, and the ability to collaborate across diverse teams. Showing a passion for healthcare and data science also sets candidates apart.
PracHub interview research ↗What is the culture and working style at UC San Francisco?
UCSF fosters a collaborative and innovative culture, emphasizing teamwork and a commitment to improving health outcomes. Candidates should be prepared to engage with a diverse range of stakeholders.
PracHub interview research ↗What is the typical timeline from initial screen to offer?
The timeline can vary, but candidates typically hear back within a few weeks after their interviews, with an offer potentially following shortly after.
PracHub interview research ↗Sources & methodology 3 sources ↗
No official company page is cited. Rounds and questions come from candidate reports and PracHub editorial material; each source shows the date it was read.
- 01PracHub interview research ↗
PracHub editorial research into this company and role, maintained with this guide. Candidate-reported, not an employer publication.
PracHub page · Accessed 2026-09-22 - 02PracHub Data Scientist practice ↗
Cross-company practice questions for this role.
PracHub page · Accessed 2026-09-22 - 03PracHub interview preparation framework ↗
The framework the preparation plan follows.
PracHub page · Accessed 2026-09-22
Explore related engineering environments and compare the preparation focus.
Veracyte Data Scientist Interview Questions & Guide 2026
Causal-inference exercises for Veracyte data scientist prep: difference-in-differences, matching, instrumental variables, and knowing when observational data simply cannot answer the question asked.
Zoetis Data Scientist Interview Questions & Guide 2026
Metric-diagnosis practice for Zoetis data scientist candidates: a number moves week over week and you work out why, splitting by segment, ruling out a logging or release change, and separating a mix shift from a real change in user behaviou
Vizient Data Scientist Interview Questions & Guide 2026
Communication practice for Vizient data scientist candidates: turn an analysis into a recommendation, name the uncertainty honestly, and answer the follow-up a skeptical stakeholder will raise.
Vibrant Emotional Health Data Scientist Interview Questions & Guide 2026
Case practice for the Vibrant Emotional Health data scientist interview, where the prompt is deliberately underspecified and your first move is scoping it into something measurable.