How do you choose the initial traffic percentage for a model canary?
PICTURE THIS: HOW TO EXPLAIN IT
Simple meaning
Start small enough that a disaster is contained, but large enough to see latency and error signal within minutes, often 1 to 5 percent.
WHY — Canary Deploy instead of guessing?
Why interviewers care about Canary Deploy:
question about Canary Deploy.
trade-offs, and what you would actually do on a MLOps project - not buzzwords.
Name the idea, why it exists, then one short example.
End with when you use it and one common pitfall.
STEPS — What happens step by step?
Before you speak the answer, walk the interviewer through these steps:
- 1Start small enough that
a disaster is contained, but large enough to see latency and error signal within minutes, often 1 to 5 percent.
- 2Increase only after quality
proxies look stable.
- 3Statistical quality metrics may
need more traffic or a longer window than ops metrics.
- 4Give an example
One tiny concrete case you can say aloud.
- 5Common mistake
What juniors usually get wrong.
- 6Close
When you pick this over the alternative.
EXAMPLE — See it in action
Here's a short line you can speak, broken into clear beats:
Note: Adapt this scaffold to your own project — keep it under 60–90 seconds.
Key takeaway
Start small enough that a disaster is contained, but large enough to see latency and error signal within minutes, often 1 to 5 percent. Increase only after quality proxies look stable.