Recap of Google Next 18
I listened in to the livestream of Google Next 18. Would love to hear from others about their perspective on the conference.
Here are the key takeaways from my perspective
- “Google Cloud sees security as a primary concern and AI as its major opportunity “ — Diane Greene .Google’s focus is on Security. Security starts at the hardware and there is encryption of data at rest and in transit. AI is a huge opportunity. Focus is on rolling out AI to the masses ( Democratizing AI ) using Tensorflow, AutoML. Google has opensourced TensorFlow.
- Google is a leader in Open Source — Examples TensorFlow and Kubernetes.
- Why Google Cloud ? — Information is driving the business. Innovation is continually required. 20 years of innovation at Google. Google has the most advanced cloud. All running carbon neutral. Hyper fast machine to machine connectivity. Fast growing body of AI resources. Security starts with the Titan chip ( checks boot by default). Named a leader in Security by Forrester. Developed verticals for the various industries — Health, Financials, Transportation, Oil & Gas etc, Gaming, Retail.
- Target is running on Google Cloud. Tech has gone from being an anchor to driver of the business. Google was important to the Engineering team and so it was important to Mike Mcnamara, CIO of Target.
- Forward looking company — They are proud of being cutting edge. Working on Quantum computing.
- GSuite’s gmail stops 99.9 % of spam.
- GKE on-prem announced.
- Integrate voice API’s into your demos using DiaglogFlow
- Google Cloud for Startups. You can apply here.
- Kaggle community — Learn machine learning quickly. Python and R are the main machine learning languages. Six hours of compute per session at no cost.
- Java script has the highest number of commits on Git ( !!!)
- Google focuses on the cloud using Open source technologies.
- Serverless does not mean there are no servers. It just means that you do not have to care about the servers in a public cloud.
- Kubernetes will become the cluster operating system in a few years. This layer gets a lot of innovation at this time. Knative and other capabilities will layer on top of the Kubernetes API’s. Kubeflow
- Talent shortage will be the biggest problem with AI. More developers need to enter this field. Google has an internal Ninja program. They will be inviting their customers to join the program. Hence they introduced AutoML.
- Applied AI has to be easy to use in the applications — Kelsey Hightower.
- Question to the panels: What keeps you up at night: Too many abstractions making trouble shooting harder. Automating the machine learning expertise.