Both Ana and Bo are noticing linguistic trends, but their generalizations aren't quite correct. This linguistic puzzle is something of a logic puzzle, and you can look for trends in the word construction to come up with the correct rules to explain the data.
Ana states that [q] occurs after back vowels and [k] appears elsewhere, but that's not the case. In the sample given, [q] occurs only after [a]. If you look at a linguistic chart that shows how vowel sounds are made, you'll see that [a] is only one of several back vowels. Other back vowels include [u] and [o] (and there are more back vowels that require specialty phonetic symbols). So, Ana's generalization is too broad.
Bo states that [k] occurs at the end of a word and [q] appears elsewhere, but this is clearly incorrect given data like [enaq] and [naq]. His statement about [k] is correct in the given sample—we don't see a [k] anywhere except at the end of a word—but his generalization about [q] is too narrow. We know from thinking through Ana's statement that [q] can appear at any point in a word as long as it follows [a].
In short, Bo is looking at placement within the word, and Ana is looking at vowel use. Both are helpful ways to understand how words are constructed in a specific linguistic data set, but neither gives a complete picture on its own.