Graded in a sandbox
Test code not shown · 35 min suggested
Run tasks with a concurrency limit
Fifty uploads, but only three at a time.
All required tests passed
How to read this: the verdict comes from the required tests. The counts below describe how the code was written. Nothing here shows who typed it. Iterations are the times the tests were run. Versions are distinct versions of the code the tests ran on. Span is the time from first run to save.
- Required tests
- 6 of 6
- passed
- Bonus
- 1 of 1
- passed
- Iterations
- 3
- graded runs
- Versions
- 3
- distinct versions graded
- Span
- 2 min
- first run to save
- Runtime
- 614 ms
- in the sandbox
- Saved
- Oct 15, 2026
How the code arrived
- First graded run
- Did not pass every required test.
- Largest single jump
- 341 bytes added between two graded runs, counting from the starter code.
Tests
- PASSreturns results in the order of the tasks, not the order they finished
- PASSstarts only as many tasks as the limit allows
- PASSnever runs more than the limit at once
- PASSstarts the next task as soon as one finishes (not in batches)
- PASSresolves an empty list to an empty array
Questions to ask in the interview
The first questions come from this Challenge. The rest come from the counts above. Each one asks the candidate to explain the code in their own words.
- How do you start the next task the moment one finishes, rather than after a batch?
- What happens to tasks that are already running when one of them fails?
- How do you keep results in the order of the tasks array?
- How would you change this to collect every failure instead of stopping at the first?
- Your first graded run did not pass everything. What did you change to get it passing?