Building a restaurant‑recommendation feature with Nearby Friends signals

Q: Building a restaurant‑recommendation feature with Nearby Friends signals

Evaluates real-time nearby restaurant recommendations using opt-in location and social signals. Strong answers cover data sources, ranking, privacy, battery, validation, success metrics, merchant fairness, and risks.

Q: How do I approach Analytics & Experimentation interview questions?

Analytics & Experimentation questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master analytics & experimentation interviews.

Q: What difficulty level is this interview question?

This is a hard difficulty Analytics & Experimentation question, commonly asked during Onsite rounds at Meta.

Q: What role is this question designed for?

This question is commonly asked for Data Scientist candidates at Meta during technical interviews.

Question

Real-time Nearby Eateries Recommendation

Meta wants to leverage real-time, opt-in location from Nearby Friends to recommend nearby eateries users might like, while balancing utility with privacy, battery, fairness, and merchant quality.

Outline the product and data approach, forecast behavioral impact, define validation and success metrics, and anticipate risks.

Constraints & Assumptions

Location use must be opt-in, transparent, and revocable.
Recommendations should be useful in real time and respect privacy.
Include offline and online validation.
Consider fairness across merchants and neighborhoods.

Clarifying Questions to Ask Guidance

What exact location signals are allowed and at what precision?
What action should recommendations drive: click, save, visit, order, share, or meetup?
Which merchants are eligible, and how is quality verified?
Are purchase, visit, or check-in outcomes available?

Part 1 - Feature and Data Design

How would you use Nearby Friends location to build the feature, and what data sets are required?

What This Part Should Cover Guidance

Use live location, accuracy, motion state, time, historical visits, preferences, social graph, merchant POI data, hours, ratings, popularity, and context.
Include candidate generation, ranking, privacy filtering, and battery-aware updates.
Use merchant and user features without exposing sensitive location.

Part 2 - Impact and Validation

Why create restaurant recommendations, how might behavior change, and how would you validate model effectiveness?

What This Part Should Cover Guidance

State hypotheses around offline visits, saves, clicks, shares, and social coordination.
Validate offline with relevance labels and historical visit prediction.
Run A/B tests with recommendation uptake, click-to-visit, conversion, retention, and satisfaction metrics.

Part 3 - Post-launch Metrics and Risks

Which metrics and negative impacts would you monitor?

What This Part Should Cover Guidance

Track recommendation impressions, CTR, saves, visits, orders, social shares, repeat use, merchant distribution, and user satisfaction.
Monitor privacy complaints, opt-outs, battery drain, location accuracy, biased merchant exposure, spam, and safety issues.

Follow-up Questions Guidance

How would you estimate offline visits without invasive tracking?
What if recommendations overexpose already-popular restaurants?
How would you handle users traveling in unfamiliar cities?

Building a restaurant‑recommendation feature with Nearby Friends signals

Quick Overview

Building a restaurant‑recommendation feature with Nearby Friends signals

Real-time Nearby Eateries Recommendation

Constraints & Assumptions

Clarifying Questions to Ask Guidance

Part 1 - Feature and Data Design

What This Part Should Cover Guidance

Part 2 - Impact and Validation

What This Part Should Cover Guidance

Part 3 - Post-launch Metrics and Risks

What This Part Should Cover Guidance

Follow-up Questions Guidance

Write your answer