I agree. Of course genes are involved... but that doesn't make it Mendelian. Autism would be caused by a combination of genetic and environmental factors.
The fact exists that there is no singular autism gene. And there could never be a genetic test for autism. It could predict the likelihood, but it wouldn't be pass/fail. Because there are many autism-related genes but no single gene trigger like a light switch. And you could still have all of those genes turned on and not have autism due to environmental factors.
Obviously, learning all popular tech stacks is not entirely realistic, but IMO when a SE or any engineer is looking for work, at least 1-2 months of researching and getting proficient with the full breadth of what's out there for your interest is going to pay off.
Unless you are moving jobs to just do the exact same thing somewhere else (who wants that) you kind of have to take it on yourself to show up the the interview appearing like an expert in skills you hope to get to really grow comfortable with on the job. You may even need to exaggerate your experience.
Ultimately, specialization should be more about how you think than what facts you know, but the hiring market won't see it that way, meaning you need to just get really good at faking it and not be afraid to back it up later after some late-night study sessions.
The other half of this is that 'levels' don't always translate across environments. Just find work that resonates with you, and you will move up. Obsessing over the level (or even the pay) is probably limiting.
The fact exists that there is no singular autism gene. And there could never be a genetic test for autism. It could predict the likelihood, but it wouldn't be pass/fail. Because there are many autism-related genes but no single gene trigger like a light switch. And you could still have all of those genes turned on and not have autism due to environmental factors.