How do I practice coding and algorithm questions?

Use PracHub's coding console to write, test, and debug your solutions in Python or JavaScript. View hints, test against sample inputs, and compare with official solutions.

What difficulty level is this coding question?

This is a medium difficulty Coding & Algorithms question, commonly asked during Technical Screen rounds at Anthropic.

What role is this question designed for?

This question is commonly asked for Machine Learning Engineer candidates at Anthropic during technical interviews.

Convert Samples into Event Intervals | Anthropic Coding Question

Quick Overview

This question evaluates understanding of array and sequence processing, run-length encoding concepts, and interval representation for time-ordered traces, including handling boundary conditions and filtering runs by a minimum length k.

Loading coding console...

Part 2: Convert Samples into Event Intervals with Minimum Run Length k

You are given a time-ordered list `samples`, where `samples[i]` is the function name observed at integer timestamp `i`, and an integer `k`. Convert the trace into a list of events `(function_name, start_time, end_time)`, but only keep a run if the same function appears for at least `k` consecutive timestamps. Use half-open intervals `[start_time, end_time)`. Runs shorter than `k` must be discarded. Important: do not merge runs that were separated in the original trace, even if the separated runs have the same function name and the run between them was discarded. Example: if `samples = ["A", "A", "B", "B", "B", "A"]` and `k = 2`, the output is `[("A", 0, 2), ("B", 2, 5)]` because the final `A` run has length 1 and is discarded.

Constraints

0 <= len(samples) <= 200000
Each `samples[i]` is a string
1 <= k <= 1000000000
Do not merge separate runs, even if short runs between them are discarded
The output must use half-open intervals `[start_time, end_time)`

Examples

Input: (["A", "A", "B", "B", "B", "A"], 2)

Expected Output: [("A", 0, 2), ("B", 2, 5)]

Explanation: The last A run has length 1, so it is discarded.

Input: (["A", "A", "B", "A", "A"], 2)

Expected Output: [("A", 0, 2), ("A", 3, 5)]

Explanation: The middle B run is discarded, but the two A runs must remain separate because they were not contiguous in the original trace.

Input: (["A", "B", "A", "A", "C", "C", "C", "A"], 2)

Expected Output: [("A", 2, 4), ("C", 4, 7)]

Explanation: Only the A run from 2 to 4 and the C run from 4 to 7 meet the minimum length.

Input: ([], 3)

Expected Output: []

Explanation: Empty input produces no events.

Input: (["Z", "Z", "Z"], 1)

Expected Output: [("Z", 0, 3)]

Explanation: With k = 1, every run is valid, so the whole array becomes one event.

Hints

First identify each maximal contiguous run exactly as in Part 1.
When a run ends, compute its length as `end - start` and append it only if that length is at least `k`.

Quick Overview