non-decreseasing order sounds better, than increasing order with possible duplicates. Maybe you could change that to make the description clearer

the constraints should be way bigger for a 4 kyu

rounded to 4 decimals

That's generally a very bad idea. Tests for all translations should use equals with delta (say, 1e-3): that will make it compatible with existing solutions, but also will get rid of the usual problems when working with floating-point numbers.

In my opinion, tests like this should use an epsilon value to check the result. This would reflect actual real world usage, especially when computing geometric values. Storing the rounded result in a variable makes no sense at all.

add dynamic programming tag imo

reference solution sometimes returns No decreasing sequence for the longest increasing subsequence method, and sometimes returns No increasing sequence for the longest decreasing subsequence method.

Kata should specify which letters are vowels. It should not be assumed that the vowels include/exclude y.

int above 20

[[[10, 10, 9], [11, 9, 9], [11, 10, 8], [11, 11, 7], [12, 9, 8], [12, 10, 7], [12, 11, 6], [12, 12, 5], [13, 8, 8], [13, 9, 7], [13, 10, 6], [13, 11, 5], [13, 12, 4], [13, 13, 3], [14, 8, 7], [14, 9, 6], [14, 10, 5], [14, 11, 4], [14, 12, 3], [14, 13, 2], [14, 14, 1]], 21, 29]
should equal
[[[11, 10, 10], [11, 11, 9], [12, 10, 9], [12, 11, 8], [12, 12, 7], [13, 9, 9], [13, 10, 8], [13, 11, 7], [13, 12, 6], [13, 13, 5], [14, 9, 8], [14, 10, 7], [14, 11, 6], [14, 12, 5], [14, 13, 4], [14, 14, 3], [15, 8, 8], [15, 9, 7], [15, 10, 6], [15, 11, 5], [15, 12, 4], [15, 13, 3], [15, 14, 2], [15, 15, 1]], 24, 31]

my solution seems better than the 'should' value.

Python:

The function name in the initial solution setup still uses camelCase:

• count_specMult should be count_spec_mult

I optimised from passing 30 test cases to 200 test cases in python and am now running out of ideas for optimization. Is it supposed to be more optimised or is the random test case is in infinite loop causing timeout?

JS fork

• updated to node 18
• made description language-agnostic
• used mocha/chai assertions + lodash for randomness
• better feedback on failure (this issue)
• rewrote random tests for more cases coverage
• got rid of reference solution