Design fast search for AV raw logs

Q: Design fast search for AV raw logs

This question evaluates a candidate's ability to design scalable logging and search infrastructure for autonomous-vehicle telemetry, testing skills in large-scale distributed systems, high-throughput data ingestion, time-range indexing and retrieval, and the trade-offs between relying on raw sensor signals versus derived perception outputs.

Q: How do I approach System Design interview questions?

System Design questions require understanding of core concepts and practice. PracHub provides solutions with explanations to help you master system design interviews.

Question

Loading...

Autonomous vehicles collect large volumes of raw logs (e.g., sensor streams, perception outputs, planning/debug signals). In these logs, there may be situations such as:

A pedestrian crossing
A traffic sign appearing (stop sign, speed limit sign, etc.)

Design a system that allows engineers to quickly find log segments where these situations might occur.

Requirements

Users should be able to query for scenarios like “pedestrian present”, “stop sign detected”, optionally with filters such as time range, vehicle/software version, geography, route, weather, etc.
Results should point to the exact time ranges (segments) in logs.
Logs are large-scale (many vehicles, continuous uploads).

Clarify

Explain what signals you would rely on (raw sensor vs derived perception outputs), how you index/search, and how the system scales.

Design fast search for AV raw logs

Quick Overview

Requirements

Clarify

Solution

Comments (0)