← DoorDash Interview Insights

DoorDash·Machine Learning Engineer·Technical Phone Screen·Senior

Senior
May 2026

Summary

DoorDash ML engineer interview with a meaty growth/promo design question. The whole session basically lived inside one open-ended scenario, so if you blank on uplift measurement or budget allocation you're in trouble fast.

Questions Asked (1)

Q1

You own a promotional offer system (e.g. '$10 off' or 'free delivery') meant to drive growth. Walk through how you'd decide who gets a promo, what offer to send them, and when to send it, given a limited budget, the risk of discounting users who'd order anyway, and the need to measure real incremental lift without tanking long-term margins.

A/B Testing & ExperimentationProduct Analytics & MetricsSystem Design
Author's notes

This question is deceptively wide.

Create a free account to read the full note

AI HintsAI Generated

Suggested Approach

Frame the problem as a constrained optimization: maximize incremental profit by targeting users with high uplift and low organic conversion probability, using causal ML to estimate heterogeneous treatment effects. Propose a two-stage system: first, offline uplift modeling to personalize offers, then online experimentation with holdouts to measure true lift and guard long-term margins.

Pro tip: Emphasize that the goal is not to maximize conversions but to maximize incremental profit per dollar spent, and that you'd use a long-term holdout to monitor cannibalization and retention effects beyond the campaign window.

1. Define objective and constraints

Clarify the business goal (e.g., incremental orders, profit) and constraints (budget, margin floor, user experience). Establish a north-star metric that balances short-term lift and long-term value.

2. Estimate incremental impact with causal ML

Use historical data to build uplift models (e.g., meta-learners, causal forests) that predict the difference in outcome between treated and untreated for each user. Identify 'persuadables' who are likely to convert only with a promo.

3. Optimize offer assignment under budget

Formulate as a knapsack or linear programming problem: allocate offers to users where expected incremental profit (uplift * margin - cost) is positive, subject to budget. Personalize offer type and timing based on user features.

4. Run randomized experiments to validate and measure lift

Deploy a randomized controlled trial (A/B test) with a holdout group to measure true incremental lift. Use techniques like switchback or cluster randomization if interference is a concern. Monitor guardrail metrics (e.g., long-term retention, margin).

5. Iterate and monitor long-term effects

Continuously retrain models with new data, adjust targeting as user behavior shifts, and maintain a long-term holdout to detect degradation in organic behavior or margin erosion. Use multi-armed bandits for adaptive allocation if appropriate.

Key Points to Mention

  • Uplift modeling / heterogeneous treatment effect estimation to avoid discounting users who would convert anyway
  • Randomized controlled trials with holdout groups to measure true incremental lift
  • Budget-constrained optimization (e.g., knapsack, linear programming) to allocate offers efficiently
  • Long-term holdout and guardrail metrics to monitor margin and retention impact
  • Personalization of offer type and timing based on user features and context
  • Potential for interference and the need for cluster or switchback randomization in marketplace settings

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