[lol768]

> Every week, you are matched with another dev using an ML algorithm for a 1 on 1 call.

I must admit I rolled my eyes when I read "ML algorithm" since it's usually marketing bullshit. Seemingly it's unsupervised knn w/ Scikit Learn, mixing variables like "open to video calls", "location", "openness to ideas" and "timezone".

Fairly neat.

[shultays]

What if it used an AI to match you instead? We can even run that AI on blockchain!

[throw1234651234]

I literally can't even, trigger warning! I hope you are using that blockchain on an ethically sourced quantum computer using ethically raised qubits.

[tr1ll10nb1ll]

It indeed is. I'll try to find someone better at ML than me over time but I think this approach should serve well at least for the first 1k users or so if I'm not wrong.

[swyx]

maybe just say “similarity” rather than “ml algorithm” since that is what you are doing

[tr1ll10nb1ll]

From the POV of a marketing copy, I don't think that sticks. I could possibly change it on the Github, however. Or, elaborate on what is meant by an ML algorithm in the context of DevClad.

[jamessb]

> you are matched with another dev using an ML algorithm

Could be simplified to just:

> you are automatically matched with another dev

[tr1ll10nb1ll]

Fair point.

[me551ah]

Isn't "unsupervised knn w/ Scikit Learn" ML? Yes it's unsupervised and not a lot of effort was put into it, but it is still ML.

[saurik]

The usage of "usually" in the comment you are replying to is often paired with a "but/except" that I feel is implied here, and so they actually agree with you.