Design the event instrumentation for Safety Toolkit at scale at global scale

Problem Statement Description

Product context: Uber is a mobility and delivery platform; its products include rides, Uber Eats, grocery and retail delivery, freight, driver and courier tools, and marketplace pricing.

Uber’s Safety Toolkit is a critical in-trip experience used by riders, drivers, couriers, and delivery recipients across global mobility, delivery, and logistics use cases. In this interview, you are asked to design the event instrumentation needed to understand how the Toolkit is discovered, opened, used, completed, failed, and trusted at global scale.

Assume the Toolkit includes safety actions such as emergency assistance, trip sharing, reporting an issue, contacting support, audio or incident-related features where available, and contextual safety questions. The instrumentation must work across regions, languages, device types, network conditions, product surfaces, and regulatory environments, while supporting both real-time safety operations and longer-term product analytics.

The focus is not to redesign the Safety Toolkit itself, but to define what data should be captured, how events should be structured, how systems should exchange and protect that data, and how product, engineering, safety operations, legal, and local market teams would use the instrumentation to make decisions. Pay special attention to price-sensitive users and markets where connectivity, device quality, trust in platforms, and local emergency workflows may vary significantly.

The experience should consider:

- Core user journeys and event taxonomy across rider, driver, courier, and delivery contexts, including entry points, intent signals, action attempts, completion states, failures, and abandonment.

- Required event properties, identifiers, timestamps, session/trip linkage, location granularity, marketplace context, and localization attributes needed for useful analysis without over-collecting sensitive data.

- APIs, client-server contracts, streaming pipelines, offline buffering, deduplication, ordering, retries, schema evolution, and data quality validation at global scale.

- Reliability expectations for safety-related telemetry, including latency, loss tolerance, fallback behavior, observability, alerting, and incident response when instrumentation fails.

- Privacy, security, consent, data retention, access control, regional compliance, and special handling for sensitive safety, location, audio, emergency, and incident-reporting data.

- Cohorts and segmentation such as geography, trip type, user role, tenure, device class, app version, network quality, affordability-sensitive markets, and high-risk scenarios.

- Product and operational use cases enabled by the data, including funnel analysis, feature adoption, safety incident triage, experimentation, anomaly detection, marketplace trust, and local operations improvements.

- Rollout strategy for instrumentation changes, including backward compatibility, phased launch, QA, monitoring, stakeholder communication, and safe rollback.

Your goal is to present a technically rigorous instrumentation design that balances user safety, privacy, reliability, and decision usefulness. The interviewer will expect clear requirements, thoughtful system and data trade-offs, and an approach that can scale across Uber’s global marketplace without compromising trust or operational responsiveness.