Production Debugging for Latency | Dataford Interview Questions

How do you approach debugging a complex issue in a production environment where latency and accuracy are critical?

For this coding exercise, implement a diagnostic scan over ordered request observations. Return each request ID once, in first-seen order, if any observation exceeds the latency limit or has different expected and observed values. The function receives a list of observation objects and an integer latency limit.

Function signature: def find_anomalies(observations, latency_limit):. Each observation has request_id, latency, expected, and observed; return a list of request IDs.

Constraints

1 <= len(observations) <= 5000
0 <= latency_limit <= 10^9
Each observation contains a request_id, latency, expected, and observed field
request_id values are hashable
Return request IDs in first-seen order without duplicates

Function Signature

def find_anomalies(observations, latency_limit):

Your question is Production Debugging for Latency. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.

Problem

How do you approach debugging a complex issue in a production environment where latency and accuracy are critical?

Function signature: def find_anomalies(observations, latency_limit):. Each observation has request_id, latency, expected, and observed; return a list of request IDs.

Constraints

1 <= len(observations) <= 5000
0 <= latency_limit <= 10^9
Each observation contains a request_id, latency, expected, and observed field
request_id values are hashable
Return request IDs in first-seen order without duplicates

Function Signature

def find_anomalies(observations, latency_limit):

Your question is Production Debugging for Latency. Start with the requirements in the Question tab.

Run and submit as often as you like. When you're ready, talk me through your approach or go straight to the code.