# Design a device fingerprinting service: system design interview question

> Estimate device matches from uncertain signals while controlling false matches, lookup latency, signal changes, and privacy risks.

Source: https://sysround.com/questions/design-device-fingerprinting-service
Updated: 2026-11-05


Difficulty: hard. Levels: Senior, Staff+.

## How the interviewer opens it

"Design a service that estimates whether a request comes from a device it has seen before. Client signals can change or be false."

## What to clarify first

- Meaning of a match
- Available signals
- Scale
- Read and write mix
- Latency and freshness
- Consistency and durability
- Match quality
- Record sizes and retention
- Privacy and erasure
- Excluded behavior

## Scope of a strong answer

Core requirements:

- Estimate a device match within one tenant and return confidence or unknown
- Record eligible observations and tolerate normal signal changes
- Measure precision and repeat-device coverage with independent labels
- Enforce consent, retention, and data erasure

Usually out of scope:

- Identify a person
- Link devices across tenants or websites
- Make final authentication or fraud decisions
- Collect hardware identifiers or invasive browser signals

## What interviewers probe

### Match quality

Many devices share the same signals. The service must improve useful coverage without exceeding the false-match limit.

### Signal changes and false reports

Normal changes and false reports can look alike. Unsafe profile updates let an attacker alter future match results.

### Lookup latency

Common signals can describe millions of devices. The service must bound lookup work without losing all useful coverage.

### Consent and erasure

Observations spread into profiles, indexes, caches, and journals. Delayed work and recovery can restore data after erasure.

## Common mistakes

- Treats a fingerprint as a unique device identifier.
- Reports high precision while ignoring near-zero coverage.
- Uses previous match decisions as independent device labels.
- Lets uncertain matches change profiles without limits.
- Ignores large lookup workloads from common device traits.
- Deletes profiles but retains recoverable evidence in indexes, journals, caches, or backups.

## The bar by level

- **Senior:** Bounds lookup work and defines a valid precision and coverage evaluation. Handles signal drift, unsafe updates, and delayed observations. Covers index history and payload keys in the erasure design.
- **Staff+:** Defines release decisions under uncertain labels and population changes. Controls tenant interference and recovery capacity. Traces erasure through concurrent writes, journals, key copies, unavailable replicas, and restores.

Practice this question with an AI interviewer: https://sysround.com/practice/new?problem=device-fingerprinting


---
SysRound (https://sysround.com) is an AI practice tool for system design interviews. It has a voice or typed AI interviewer and a shared whiteboard. At the end, you get feedback on each skill and a model answer. The AI feedback can be wrong.