The Use of Botting

Edge of the Falls

This article on the use of bots on github made me think of a different use of the github api.

The first pieces of python code I pushed to github on my own account were in my tiny-artnet mircopython artnet implementation. Soon after committing that code I started getting emails from recruiters looking to hire python developers. They would say something along the lines 'based on your github activity we think you would be perfect for a job doing django".

At first these were hilarious, micropython is nothing like python, if they had looked at my github profile they would have seen the large C projects I work on.

But after a few of these I started to get annoyed, clearly these people were finding my email from code I had written or from commit logs. Why weren't they trying a little bit harder? To me, github is the technical recruiters wet dream, but whoever was generating the leads here clearly wasn't doing a good job.

I don't think cold lead generation is a good way to sell anything, let alone a job opportunity, but this is how I would use github(bitbucket, gitlab and everything else too) to do it.

  1. Search projects that have the correct language keywords (python, go, c)
  2. Find any email addresses at all, sort by most recent
  3. Attempt to resolve email addresses into real people
  4. a) Find personal site for email address or b) (worse) find social media pages for address
  5. Send generated lead info to recruiter

The human at the end needs to be able to do a final set of filters, but anywhere that is too high a cost isn't going to use the lead well anyway. I am sure the 100 line script that could be written on those lines that would generate substantially better leads than cold contacting any email address.


Reading: Reamde