← Uber Interview Insights

Uber·Data Scientist·Onsite - Behavioral / Leadership·Senior

Senior
Jun 2026

Summary

Prepped for a data scientist role at Uber and ran through a set of behavioral and technical questions that felt pretty representative of what they actually care about. The dynamic demand project question was the one I was most nervous about since it requires you to hold a lot in your head at once.

Questions Asked (3)

Q1

Tell me about a time you went above and beyond what was expected of you.

Adaptability & AmbiguityCross-functional Alignment
Author's notes

Solid question to prep but easy to fumble if your example is too vague.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Use the STAR method to describe a specific project where you voluntarily took on additional responsibility beyond your assigned tasks, focusing on how it benefited stakeholders and cross-functional teams. Emphasize the impact of your actions on the project's success and how it aligned with broader organizational goals.

Pro tip: Choose an example where going above and beyond involved influencing or aligning multiple teams, as this demonstrates stakeholder management and cross-functional collaboration—key for ML roles at Google. Quantify the impact to show measurable value.

1. Set the Context

Briefly describe the project, your role, and the expected responsibilities. Highlight the cross-functional nature of the project and the stakeholders involved.

2. Identify the Gap

Explain the additional need or opportunity you noticed that was beyond your scope, such as a missing feedback loop or an unaddressed stakeholder concern.

3. Take Initiative

Describe the actions you took to address the gap, emphasizing how you collaborated with other teams or stakeholders to implement a solution.

4. Highlight the Impact

Quantify the results of your efforts, such as improved model performance, reduced time-to-market, or increased stakeholder satisfaction.

5. Reflect and Connect

Summarize what you learned and how it demonstrates your ability to drive cross-functional alignment and manage stakeholders effectively.

Key Points to Mention

  • Cross-functional collaboration with teams like product, engineering, or data science
  • Stakeholder management: identifying and addressing needs of different groups
  • Initiative and ownership beyond assigned tasks
  • Measurable impact on project outcomes or business metrics
  • Alignment with Google's culture of innovation and collaboration
  • Learning and growth from the experience

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q2

Describe a situation where you disagreed with your team, the outcome didn't go your way, and what you took away from it.

Conflict ResolutionAdaptability & Ambiguity
Author's notes

This one is sneaky because they want you to actually fail, not just 'disagree and then be proven right.' I've seen people (myself included) try to dress up a success story as a disagreement and it reads as evasive.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Choose a real disagreement where you advocated for a data-driven alternative but the team chose another path, and the outcome was suboptimal. Focus on how you handled the loss professionally, supported the final decision, and extracted a concrete lesson that changed your approach. Show self-awareness and growth, not blame.

Pro tip: Emphasize that you committed to the team's decision once it was made—interviewers at Uber value strong opinions loosely held and the ability to disagree and commit. Then highlight a specific process improvement you adopted afterward, like running a pre-mortem or building a lightweight experiment to test assumptions earlier.

1. Set the context and disagreement

Briefly describe the project, your role, and the specific decision where you disagreed. State your position and the team's position clearly and neutrally.

2. Explain your reasoning and how you voiced it

Share the data or logic behind your view and how you communicated it—e.g., in a meeting, via a written memo, or with a prototype. Show you advocated professionally without being combative.

3. Describe the outcome and your reaction

Explain that the team decided against your recommendation and what happened as a result. Highlight that you supported the final decision and contributed to its execution, even if results were mixed.

4. Extract the lesson and show growth

Articulate what you learned—about influencing without authority, decision-making under uncertainty, or when to push vs. let go. Give a concrete example of how you applied this lesson later.

Key Points to Mention

  • A specific, data-driven disagreement (e.g., model choice, experiment design, metric definition) relevant to data science.
  • How you communicated your dissent constructively, using evidence and respecting others' perspectives.
  • Your commitment to the team's final decision and your role in executing it well.
  • The actual outcome—be honest if it was suboptimal, but avoid saying 'I told you so.'
  • The concrete lesson learned and how you changed your behavior or process afterward.
  • Growth in collaboration, influence, or decision-making that makes you a stronger data scientist.

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.

Q3

Walk me through a project you worked on involving dynamic demand forecasting or a similar problem. Cover the business context, your technical choices, any tradeoffs you made, how you worked with stakeholders, and what the actual impact was.

Technical Trade-offsStakeholder ManagementProduct Analytics & Metrics
Author's notes

This is the one that separates prep from real experience.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Choose a project where you owned the end-to-end forecasting solution, and structure your answer to highlight the business problem, your technical decisions, tradeoffs, stakeholder collaboration, and measurable impact. Emphasize how you balanced model complexity with business needs and how you communicated results to non-technical partners.

Pro tip: Quantify the impact in terms of business metrics (e.g., reduced forecast error by X%, saved $Y, improved operational efficiency by Z%) and explicitly state the tradeoffs you made (e.g., accuracy vs. interpretability, latency vs. scalability).

1. Set the Business Context

Briefly describe the company, the problem, and why demand forecasting mattered. Mention the scale (e.g., number of markets, SKUs, or time series) and the stakeholders involved.

2. Explain Your Technical Approach

Outline the data sources, feature engineering, model selection (e.g., time series, ML, deep learning), and validation strategy. Justify why you chose that approach over alternatives.

3. Discuss Tradeoffs and Decisions

Highlight key tradeoffs such as accuracy vs. interpretability, model complexity vs. maintainability, or real-time vs. batch processing. Explain how you made decisions and any compromises.

4. Describe Stakeholder Collaboration

Explain how you worked with product, engineering, operations, and business teams. Mention how you gathered requirements, communicated progress, and incorporated feedback.

5. Quantify the Impact

State the measurable outcomes: improved forecast accuracy, cost savings, revenue increase, or operational efficiency. Tie the impact back to the original business problem.

Key Points to Mention

  • Business problem and why forecasting was critical (e.g., supply-demand matching, inventory optimization).
  • Data sources and feature engineering (e.g., historical demand, seasonality, promotions, external factors).
  • Model selection and validation (e.g., ARIMA, Prophet, LSTM, ensemble methods; cross-validation, backtesting).
  • Tradeoffs made (e.g., model interpretability vs. accuracy, latency vs. scalability, automation vs. manual override).
  • Stakeholder management (e.g., aligning with product managers, engineers, and business leads; regular updates).
  • Quantified impact (e.g., reduced forecast error by 15%, saved $1M annually, improved on-time delivery by 20%).

AI-generated suggestions, not part of the candidate's original notes. May be inaccurate — verify before relying on them.