Wednesday, February 4, 2015

Mexico And Machine Learning (The long due rant)

Spanish Version Follows

Go to your favorite search site (Bing, Google, Yahoo) and do the following search:

"Machine Learning Jobs", Mexico

Ready? No? I'll wait a bit longer........Ready? Cool.

As you can see, the job offers in Mexico that includes Machine Learning as a keyword are from 1 to nothing. This means that no one in Mexico thinks that analyzing the data they have is worth their money.

Telmex, the largest telephone company in the world is also one of the less developed, just try putting this search terms:

"Telmex", "Machine Learning"
"ATT""Machine Learning"
"NTT""Machine Learning"

These are the largest communication companies in Mexico, USA, and Japan. While ATT and NTT have research labs Telmex does not have a single research lab oriented to data analytics.

Let me remind you, Telemex is owned by the richest man in the world, and its net profit is well over ATT's net profit, yet, no commitment to research.

So, what can we expect if the largest company in Mexico doesn't do research at all. You end up with a plethora of start-ups (Pyme) that do not care about R&D either. Even thought Machine Learning is relatively cheap compared with other R&D overhead costs, they see no interest on analyzing the data.

It hurts me to see how most of the most innovative start-ups in Mexico are basically rediscovering what has already been done in the US 10 years earlier. How, due to the negligence to Science and Technology, most people that have PhDs have either to stick to Academia or leave the country altogether.

Not like scoring a job in Academia in Mexico is easy either. The process is extremely murky, and the whole establishment is extremely oligarchic.

When I tried applying to a Job at Mexico's top University, and after sending all my documentation, their answer was: "Thank you, for your application, but be aware that is extremely difficult to get in here, so do not have high hopes"......

Who does that? What kind of self-named world class institution even has that kind of response to job applicants? Hell, I've applied to Google and Microsoft and never got that kind of answer.

And all of that brings us to Machine Learning, since is really hard to score a Job in Academia, and there is virtually no R&D, new trends (like Machine Learning) take a real long time to enter Mexico, researchers in CS are still using decades old techniques and couldn't care less about doing state of the art research because..... tenure?

Anyway, that is how things are, luckily I know plenty of people who still have high hopes, and are trying to push innovation in a real sense.

1 comment: