Good news! We've just launched JobMappr, a visual way to explore over 40,000 job opportunities! Check it out! It lists all sorts of jobs, from programming jobs to fashion jobs, retail jobs to government jobs, farming jobs to investment jobs, and so on.
It's a much more interesting way of exploring what's out there around where you'd like to work, as you get to picture yourself in all sorts of different roles and ponder what your life would be like in all those various places. It's also really interesting to see the various regional trends, to see that there are lots of jobs at financial firms in New York, lots of retail and defense contractor jobs in the Midwest, things like that. Dive in and see what you can find!
After building our Tech Jobs Map, it was only natural to generalize the idea. JobMappr uses a variety of different data sources to find as many job opportunities as possible for all sorts of people. Jobs posted on Gun.io will now also appear on JobMappr, as well on the Gun.io Tech Jobs Map, website, in the usual notifications, etc.
I was recently turned onto Amy Hoy's program, 30x500, which goes against the common notion that startups need to find a niche markets and dives deeper into how to build products with a broad appeal.
Gun.io has a fairly small niche. Relative to the overall population, only a small percentage of people are software developers, and of that percentage, only a small percentage of those are open source developers, and only a small percentage of those are looking for jobs at any one time. It's something I still think is worth doing because the way that people like me find programming work could be greatly improved, but it's still a small niche.
JobMappr has a much broader use case. Right now, the US unemployment rate is 8.2%. Multiply that by the US population, and you've got 25.3 million people who could possibly use this service. Not bad. Of course, the quality of the job opportunities which are scraped for JobMappr are a lot lower quality than the high-quality careers we list on Gun.io, but there are still quite a few good opportunities there as well.
The fun part of this project was getting the responses to be as fast as possible. The first version took over 60 seconds to load the page, and the initial query was over 14M of data! Not good. So, instead of rendering the entire database as a template and passing that to the client, the data is now separated from the code on the server, and all of the data is passed asynchronously. Additionally, it makes heavy use of Memcached, which greatly improved the response time. Now, the data points load in under two seconds, even faster than the map tiles. Hooray!
This project is still hosted on my hobby-box, so I'm not sure how well it will scale. If there is a sustained interest in the product and performance issues crop up, I will move it into a larger and faster dedicated server.
There is still some work to be done. The robots are adding more and more data all the time (a few thousand a day), but more importantly the data is quite noisy and difficult to navigate.
To solve this, I'm actually going to use some of what I learned in school for once, (yay!), to cluster the tags with the k-means algorithm. I'll try to make it as generic as possible and open source it, if I can. More on that as it happens!
Want to know more about JobMappr? Get in touch or leave a comment below! We'd love to hear any ideas you have for features and so-forth.
We match top developers with high-quality freelance and full-time jobs.
Sign up with GitHub and we'll match you with great gigs based on your open-source portfolio!
Pssst.. do you want to write for Gun.io? Or, do you have a blog you'd like to see syndicated here?
Get in touch!