There are some decent 365nm LEDs on mouser with a 3535 XP type footprint, price is still double what you paid for these though. Specifically talking about the Lite-on 365nm UV and the Luminus SST-10 365nm UV.
Those cheap 365nm leds from aliexpress are between 365 and 370 nm, so close enough. The cheap features are that the output is ok but not great, and that they produce a white/yellow-ish waste light as well, that obscures the fluorescence that you usually want the UV for. Nichia 365nm leds are the best for producing almost no visible light.
A high output and fairly cheap 365nm led (more 365nm light than most Nichia’s) that does produce some visible light is the LiteOn 365nm led (available at Mouser and tme.eu). The visible light can be filtered out with a ZWB2 filter.
Btw, the plastic optic from the DQGtiny26650 likely absorbs a lot of 365nm light, so it was wise to go for 395nm. 
I was aware that some optics can interfere with the uv but i’m not fully versed with what goes on.
Normal TIRs will be absorb most 365nm and get burned
i had an itch for high power uv, build this one.
I also have the same question.
Do those cheap anti-UVA/UVB ‘polarized sunglasses’ (many such sunglasses/night vision glasses in Banggood, for instance) offer some protection against UV light?
depends a bit on what wavelength. The 395nm light from the light in the OP needs yellow /amber glasses, the 365nm leds can be be blocked quite well with common polycarbonate glasses (most ‘impact’ glasses used in workshops and labs are made of polycarbonate).
A quick test is always: shine the UV-source onto a sheet of copier paper, all UV sources should make it fluorescence bright blue. Then place the glasses in between light source and paper and see how much fluorescence is leftover, good glasses make it almost gone.
I use uv strips on my boat for night fishing and some of the strips from China last about a month don’t know about individual leds anf how they will hold up in long run.
I used a cheap $5 pair of yellow-lensed safety glasses from Wal Mart with my 395 UV light and they worked very well. They effectively blocked the majority of the visible blue making fluorescence pop much more vividly. Even dark items were more dramatic, such as leaves that fluoresced red, and tiny mushrooms and other fungus that was washed out in the blue. They also passed the paper test Djozz mentioned; shining through the lenses would eliminate fluorescence, and my eyes could feel the difference too.
It’s close enough, just like djozz said. I bought a whole bunch of different wave lenght uv leds from there from 365 to 410 so I know the difference they make. Only the 365nm led was able to match the performance of my mini uv cfl lamp when checking currency.
The case with aliexpress is that everything depends on the seller. Some are genuine while most others are not. Remember Convoy and Soshine have their official stuff sold through Ali.
Oh crap, I didn’t know this, was planning to use mine with a TIR.
PMMA TIR wont work there was a picture here in BLF that showed a ruined triple optic and LEDs were damaged as well
PMMA is a nice base material for deep UV protection glasses below 375nm
Polycarbonate glasses absorb anything below 395nm
Wow, so I’m really just lucky I used 400nm and not 365nm 
it sure does work, i’ve build few uv lights using pmma tir lenses. have leds run for hours, no issues whatsoever.
i thinki know what picture you talking about, however it shows carclo triple, which is made of polycarb.
has nothing to do with luck, polycarbonate lenses are ok from 400nm, and up, never had any uv led damage lenses, but than i never drive them beyond reccomended current, in most cases i drive them with even less current, uv leds are not nearly as tolerant to heat as white leds, so no overdriving, 365nm will be fine with pmma, done it more than once.
I just tried the polycarbonate (clear safety glasses) and paper trick. It did reduce the fluro pop some. I then tried with/without the optics and there didn’t seem to be much difference. Does that mean I could’ve used 365nm without issues?
paper will glow at basically any wavelength, even blue light makes it glow a bit. find something else to test, id, or money, or something else that reacts to 365nm, but not to higher band.
also, are you sure your tir is not a pmma one? such 7up lenses are made by different brands, and some no name brands. also pmma is basically an acrylic plastic just not as brittle, acrylic does let lower than 400nm thru. simple common Plexiglas is acrylic.
I don’t follow?? Why do I need to test with 365nm specific colours when I have 400nm?
if you have 400, then forget about testing it, it will work with any lens\tir
Aah, so my question remains unanswered.