>I, as a non-professional in the field, have no business diagnosing anybody based on casual interactions.
That's fine, because professionals don't diagnose people officially based on "casual interactions" either, nor they do that at parties.
They have systematic scripts to go through and structured interactions. That's because diagnosing is official business and must have a higher standard plus tests and paperwork to cover your ass.
Ignoring that, whether a professional or not, if you know what to look for, it's pretty easy to tell if someone is autistic by talking to them at a social context, with way better than chance accuracy (meaning aspie autistic - for asd 2 or 3 it's way more self-evident than that even).
More often than not, the subjects reveal their diagnoses themselves at a later point.
If you have guessed correctly from before the reveal for many different individuals, you can pat yourself on the back.
For those that don't reveal or might not even have a diagnosis or be aware they could be, a pile up of additional (unrelated to the initial impression) diagnostically consistent behaviors and mannerisms as you get to know them over time is also a good enough confirmation for use outside of a clinical setting.
In general, even the average non-trained or unfamiliar with the specific traits neurotypical person is good at this identification even subconsciously, they just don't know what exactly they're identifying (so pit it as "weird", "offputting" etc):
(...) across three studies, we find that first impressions of individuals with ASD made from thin slices of real-world social behavior by typically-developing observers are not only far less favorable across a range of trait judgments compared to controls, but also are associated with reduced intentions to pursue social interaction. These patterns are remarkably robust, occur within seconds, do not change with increased exposure, and persist across both child and adult age groups
"… a static image was sufficient for generating negative first impressions of those with ASD. (...) In contrast, first impressions of TD [typically developing] controls improved with the addition of a visual information, suggesting that unlike the ASD group, visual cues helped rather than hurt the impressions they made on observers."
That's fine, because professionals don't diagnose people officially based on "casual interactions" either, nor they do that at parties.
They have systematic scripts to go through and structured interactions. That's because diagnosing is official business and must have a higher standard plus tests and paperwork to cover your ass.
Ignoring that, whether a professional or not, if you know what to look for, it's pretty easy to tell if someone is autistic by talking to them at a social context, with way better than chance accuracy (meaning aspie autistic - for asd 2 or 3 it's way more self-evident than that even).