Kaskhoruxa was one of the hardest things for me to make. At the time, I was crazy-busy with university classes, among other things, and didn't know anything about linguistics. Heck, at the time of writing this, I still don't feel I know much about linguistics.
I'm mostly self-taught and had to reinvent the wheel a lot. I remember discovering, through my own observation of English, that sentences could have transitive and intransitive verbs, as well as subjects, objects, and indirect objects. At the time, I have named these "directional verbs", "static verbs", "subjects", "targets", and "tools", respectively.
Everything changed when I joined Reddit, and found a community of conlangers. Suddenly, linguistics became quite a bit more accessible. I had also started watching Tom Scott's series on linguistics, which I found endlessly fascinating.
Another thing that I "discovered", while starting out, was the idea of "schools of sound". The idea was that while a language might have a phonology (what I called an "aesthetic" at the time), that words would not often take from the entire phonology. I'm not quite sure what this is actually called in linguistics, but I saw that for any given language, you could classify words into "schools", which each take from different parts of the phonology at different amounts. Therefore, a language would have one unified phonology, but would see internal variety where the words pulled from three to five schools of sound.
Kaskhoruxa actually started out as an English cipher, where each set of a massive list of phonemes would correspond to another set of phonemes. The idea I had at the time was that I could have words in Kaskhoruxa which would vary in syllable count from their English counterparts, and I could try to carefully balance the sound frequencies in Kaskhoruxa with the right conversion table. What I quickly realized, though, was that English has taken words from a LOT of other languages, which was what probably inspired my "sound school" idea. The issue, then, is that Kaskhoruxa would have a completely whacked set of words because the conversion table was built for only one etymological category from English. This is what I sort of considered as my "first protolang" stage, even though it's not a protolanguage, because it's a cipher. However, some words from this cipher have actually remained, though they were adjusted slightly to fit the new language.
The next stage of the language was before I really had the biological specifics down for the Avians. Because of this, I had labial sounds in the language, even though the Avians had beaks. After realizing this, I tried to write in how the beaks where an outer shell for a pair of lips underneath. However, after about of year of going along with this idea, I dumped it completely. This was also the first stage where I started implementing my own grammar, and began creating words without a cipher. My song "Rohsorbna Hruva" is actually written in this stage of the conlang, and the entire rest of the album is actually based in the really early version of the Kazhard Project, back when it was a lot more like Star Trek. I wound up revising the biology to get rid of the lips-under-beak idea, and trashed most of the conlang. Only maybe 15% of the words in this second stage evolved into the current stage of the language, which is why the Losoxulka Accent exists, as it actually makes use of a lot of these older words (though adjusted to fit the current phonology). I made this accent a thing because the second stage of the conlang had actually been worked on for so long, that I had memorized a lot of it, and it was difficult for me to shake all of the old habits for quite a while.
During the third and current version of the conlang, I had actually retained almost all of the grammar, only making a select few major changes. However, I spent a lot of time planning out how I wanted to reinvent the language, because I had just trashed the second stage, and had a chance to make some foundational changes again. One thing I really wanted to do was change the phonology in a few significant ways, mostly because I hated how it sounded in my song lol. So I was in the middle of moving out of my old house, except my mother and I had to vacate before my classes were finished at the local community college for the summer. So I spent about a month with her in a small studio-apartment-style hotel room. I was watching a video discussing how the Matoran language from Bionicle was apparently a full implementation, or extremely-close copy, of the Maori language, and the Maori people had taken LEGO to court, on the basis that they were copyrighting entire sections of their native language under the rebranded name of "Matoran". It was then that I realized that the sound that Maori had was actually closer to the sound I was trying to achieve with Kaskhoruxa, compared to many other languages I was looking into at the time. This decision formed the foundation, but did not remain for long, and now I've had some people comment that it sounds like Russian, Japanese, or German(?).
The thing was that I didn't want a language which sounded "Klingon" or "Elvish"; I wanted a language which sounded like it was sturdy and had some energy and power to it, but still being just a little muted. I guess if Klingon was the creepy warrior wearing armor of bones, while Elvish was the artist wearing long draping robes, then Kaskhoruxa was supposed to be a rusty cargo truck with some tribal geometric patterns spray-painted along the sides. Old, but not mysterious; sturdy, but not aggressive. As an added bonus, because I was doing away with labial sounds this time, I would actually have an easier time pronouncing it, than I do my own native language of English. The muscles in my lips don't move well or cooperate, which means I have a hell of a mumble problem, and have great difficulty forming words which contain "M", "B", and "W".
A weird side-effect that I sort of embraced, too, was that my first conlang, like those of most creators, would be both heavily influenced by my native language, and would be incredibly regular. When people start out, they're usually unaware of all the idiosyncracies and details that natural languages have. If they are aware, then oftentimes, people try to add a lot of regularity, because they think they're "fixing the problems of their native language". However, in the Kazhard Project story, Kaskhoruxa actually is a constructed language, made by the Ajokona. Therefore, having regularity sort of made sense.
However, the other obstacle I didn't avoid (out of ignorance, at first) was that I only spoke English, German, and Esperanto at the time, and so the grammar in Kaskhoruxa was heavily based on those languages for a long while. When I started really getting into linguistics and the conlang subreddit, I was exposed to a whole world of new and strange grammar mechanics. It occurred to me that, compared to Indo-European languages, Kaskhoruxa wasn't very alien. However, I did have a thought then, which I stand by now. That is the idea where the statistics of independently evolving a language fundamentally different from English in every way is just as unlikely as finding a language which mimics English in every way. A language could independently evolve which coincidentally shares a lot of traits with English. It's unlikely, but not impossible.
Additionally, the more I thought about it, I realized I wanted to make Kaskhoruxa ride the line between alien and familiar (to my target English-speaking audience), because the whole point of the Avians was to have aliens that weren't very alien at all. They weren't quite Star Trek's people-with-prosthetics, but they were a hell of a lot more familiar than Blares or Tauna were. I needed there to be a First Contact which Humans could approach with an outcome of peaceful resolution in their statistical favor. If Humankind is faced with an alien race, which is so completely different that fundamental ideas could result in grievous political errors, then hostility is likely guaranteed. At least, this is my line of thinking. Therefore, Avians needed to be just similar enough that Humans wouldn't be likely to stumble into some stupid social catastrophe. I didn't want my story to be another retelling of "Humans and aliens are at war and the universe is inherently out to kill us", because most of the science fiction I was exposed to portrayed that. (Remember that I was so busy working on projects as a kid, teenager, and young adult, that I didn't really have much of a chance to read a lot of other science fiction, outside of the retro stuff my father had.)
Also, one may think that they would send Humans who are specifically-trained to be extremely perceptive of social cues and be observant and practical with outsiders, but let's also remember that I'm a person with serious social difficulties, and often feel like an actual alien every waking second of my life, and people who are supposedly the super socially-perceptive ones consistently and constantly fail to interact with me, and misinterpret every other thing I do, and baffle the crap out of me on a daily basis, and this is with people who are professionally-trained to handle wide varieties of personality types, with whom I share a common spoken language. Therefore, forgive me if I don't exactly have faith in the idea that one can train a Human to forget all of their biases and nuances, and perform under a clean slate for really foreign alien travelers, and successfully pull it off. Hell, Avian similarity as it is, Humankind would still have incredible challenge getting communication and body language figured out with them. No need to stack the deck against the Human researchers on the First Contact site, lol.
Therefore, because the faction, which has First Contact, is culturally Indo-European (and so is my likely readership) then I decided to make egregious use of unlikely statistics to make Avians fairly similar to Humans, and have their language take grammatical inspiration from Indo-European ones, with a couple dozen core non-Indo-European traits tossed in for good measure, just to make sure the Avians are still very alien, similarities aside.
The fun side-effect of this, then, is that Kaskhoruxa and Avians both basically become "My First Aliens" for people exploring the Kazhard Project, where they have something that they can easily relate to, and the species and language become a stepping stone for exploring the weirder sides of the project. I've also noticed that in creating Kaskhoruxa, I've learned a lot about grammar, and in learning it, I've exercised my brain in adopting more languages and other grammars. While it's true that someone could get more utility stretching their brain using real languages, all from our very diverse and beautiful cultures on our actual planet, it's also true that I've seen many people who get super excited over Dwarvish and Elvish, while their interest in Finnish or Burmese is maybe passing, at best. If the fact that this language, made for an alien species in a story which they enjoy, is what stretches their brain and acts as a learning opportunity for exploring other languages, then I still think it has utility. After all, it has some Indo-European qualities, but it's not as Indo-European as Esperanto or German.
And while, true, designing a conlang to be accessible to a fandom may be a poor use of resources, because few fans learn the languages, I still want to do this for the few people who do decide to, because that's my gift from me to that specific part of the fandom. I'm not making this for a company or for profit; I can take the time to detail things like this. Also, frankly, many of the people, so far, have followed this Project specifically to learn Kaskhoruxa, as of writing this. There aren't many people following this right now, though; maybe 8, but 6 of those 8 expressed that they came to the site to look at Kaskhoruxa, and 4 of those 8 are frequently asking me when I'll be posting lessons for the conlang, because they really want to learn it. And those 8 people are why I've been working on this massive story since around 2014, despite it just leaving the planning stage in mid-2018.
And so those are the experiences and thoughts I had on bringing this conlang to the Project. For alien languages, which are not reminiscent of Indo-European ones, there will be plenty of strange things to find once I get properly started in Blare and Tauna language family.