COMMENT | There’s a serious problem facing the channel. Technology is moving faster than the job market. Breakthroughs in biztech are moving from concept to deployment before training institutions have produced enough graduates to manage them.
This is most evident with algorithms. The potential of a well-architected algorithm to improve efficiency in a business is enormous. I spoke to a data scientist working for a big four firm last week. He had just come back from an engagement with a very large mining company (which he couldn’t name).
That mining company had problems with payments going to the wrong places. In two weeks this data scientist created an algorithm, trained it on large amounts of data, and put it into production. It saved the mining company $70 million.
Algorithms are just as useful for SMEs. And not just with financial transactions but with any type of data that needs to be categorised or sorted. For example, it could be inventory, sales or transport data. A good algorithm can take that data and turn it into highly-valuable recommendations to grow a company.
It’s not just that there are too few graduates but the fact that they lack commercial experience. These newly minted graduates know how to build algorithms but they lack all the nuances to make them work effectively and get the right result. There is already enormous competition to hire data scientists and data engineers who have spent at least a couple of years in the commercial world. They earn salaries that would put them out of reach for most resellers.
This problem is going to happen all over again with blockchain. We have had the hype and prototypes. Now we’re seeing the first mass market products.
At Google Cloud Next, the developer conference in San Francisco last month, Google released Digital Ledger. This is an in-house blockchain that is built to scale. It has been designed to run on google’s AI platform Tensorflow. Tensorflow is named because Google has built data centres that use processors optimised for machine learning.
Digital Ledger is a database that provides a permanent, tamper-proof method for storing data. This has obvious immediate applications with financial data, as you could use Digital Ledger to keep a permanent record of all your bank transactions. If you wanted to change accounting programs you would export all the transactions and off you go.
This will become even easier in July next year when we move to open banking. The banks will then be obliged to give all your banking data to you or anyone else that you approve. And they will no longer be allowed to charge for it.
You can also store any type of data in Digital Ledger – and connect it to Google’s AI services to run algorithms you have built yourself. While everyone including Amazon and Microsoft push their own AI services, Google has one very interesting differentiator, still in alpha. Google’s Cloud Auto-ML automates the ongoing work of a data scientist by auto-tuning the algorithm.
You will still need a data scientist to set it up but you no longer need to ensure the algorithm keeps improving from its launch state. Google’s platform does that for you.
Now you may be thinking – what do blockchains and algorithms have to do with me? I’m an MSP! The short answer: the secret to a thriving business is staying relevant to your customers.
If you’re just providing infrastructure services (whether network, compute, devices, etc) then you’re leaving the door open for someone else to waltz in and capture your customer’s attention.
Blockchain and algorithms. Two critical trends you need to know about. If you’re not reading and thinking about how they could help your clients then you could be blindsided by what happens next.
Sholto Macpherson is a journalist and commentator who covers emerging technology in cloud