How Google is Remaking Itself as a “Machine Learning First” Company
Steven Levy
2.3K61

Google is retraining its coders into ML specialists because Google is smart. Anyone with an intuition on ML knows that manual coding is passe’. The digital sands are shifting at a seismic scale: computers are good at programming, humans are not. Procedural programming is monkey work, a robot’s job. Anyone coding for a living, and who is *not* developing a machine learning model, is cycling without a chain.

The reality is this: Programmers are currently weaving together software from an exponentially increasing library of inter-dependent “threads” in order to build more efficiently: The resulting fabric however becomes more and more fragile as any one these threads fails (re Kik). The situation is untenable. Programmers are confronted with a physically painless version of Dr Benway’s switchboard. Only a machine can operate this switchboard of interdependencies sanely.

As a UX designer I see enough “smart” people simply burying their heads in the sand and continuing business as usual. This works in the short to medium term — as the cogs of change churn — but realistically we are all simply treading water. It’s time for business owners, and even governments alike, to realise that it is not business as usual. All resources are to be diverted to avoid the oncoming iceberg safely. Any efforts not directed toward a robot driven world, be it physical robots or artificial intelligence text/voice bots, is a distraction from the oncoming reality.

Bots are coming to occupy plenty of “jobs”. The “jobs” that will be taken are the ones nobody actually wants to do. So this is a good thing. Bots will destroy any and all mechanised, manual tasks of a repetitive nature. The identification and elimination of duplication and redundancy is software’s job.

I say to all product owners, business leaders and Governments. It’s time to simply wakeup and start aligning towards the new world as a priority task. This is not business as usual.

Derry