Design a face recognition system
PremiumML systemMedium
- Face recognition
- Face embeddings
- Liveness detection
Quick answer
This ML system interview question asks you to design a face recognition system. Aim for a design that turns data into useful predictions in production, then explain the data, training, serving and monitoring. Interviewers want to see how you frame the problem, measure success and keep the model healthy after launch.
The problem
Design a face recognition system for office entry: employees who opt in walk up to a gate, a camera recognises them among the people allowed at that site and the gate opens. The hard parts are near-zero false matches, fair accuracy across skin tones, ages and genders, and handling biometric data with consent.
Employees who do not opt in keep using their badge, with no penalty. High-security rooms still need a badge tap together with the face check.
What must a face recognition system do?
- Enrol: An employee gives written consent and enrols 3 photos in the app or at a kiosk; the system stores a face template, not the photos.
- Identify at the gate: Match a face against the templates of people allowed at that site and open the gate.
How well must it do it?
- False matches: At most 1 in 100,000 attempts opens the gate for the wrong person.
- Misses: At most 1% of enrolled employees are not recognised and must badge in.
What scale should you design for?
Use these numbers in your estimates.
The scale numbers unlock with this questionUnlock
Key concepts
- Face recognition
- Face embeddings
- Liveness detection
- Biometric consent
- Demographic bias
- False match rate
How you’ll practise
- Problem framing
- Data and labels
- Features and model
- Training and evaluation
- Serving and monitoring
- Wrap-up
Lala, personalized to you, gives a hint when you ask and reviews your design when you are ready.
What do interviewers look for?
In any ML system design round.
- Frames the product goal as a clear ML problem
- Defines offline and online success metrics
- Explains where the data and labels come from
- Separates training from serving and keeps features consistent
- Plans for drift, retraining and rollbacks
- Weighs model quality against latency and cost
Questions people ask about this interview question
- Which design type is this question?
- It is an ML system question. You design a face recognition system and explain your choices the way you would in an interview.
- Is this question asked at entry level?
- It is marked Medium. Entry-level engineers can practise it, but it expects you to reason about requirements and trade-offs, not only list components.
- What should I cover first?
- Start with the product goal, how success is measured and what data is available. Agree on scope with the interviewer before you design anything.
- Can I practise it for free?
- It is a premium question. Sign up and unlock it once for 10 credits; it then stays unlocked. When you finish, Lala, your personalized AI mentor, can review your design for 30 credits.
Related questions
Updated · Written by the SevyDevy team