Just back from a busy day at AWS Invent 2018 where, in case you’ve had your head in the sand for the past 12 months, the buzz was all A.I. It’s just a shame that Amazon didn’t use a little machine learning input in planning timings and logistics. 10 minutes really wasn’t enough time to get from one talk to another and separate bandwidth for presenters and the punters might have saved some blushes!
Minor gripes aside there were lots of talks and workshops introducing the core AWS services including Sagemaker for AI, Alexa Skills, Greengrass for IoT, Neptune for graph database work and of course the ever present Lamda for the belle de jour – “Serverless” backends.
The AWS team has done a good job of packaging up AI workflows into SageMaker and providing access to plenty of standard algorithms as well as enabling users to add their own algorithms. Uploading models is a snip and being able to select your hardware based on training vs consumption modes is a good touch. There were a number of sessions covering working with SageMaker including an introduction to getting the Fast.ai algorithms and examples up and running.
I found the talk on using Neptune to build out a recommendation engine insightful. The background on the different query languages available (Apache TinkerPop Gremlin and SPARQL) was well explained. The working example of enriching a simple book database search with related concepts (place) was really clear.
The sheer size of the event spread across 3 floors and several sections of the London Excel centre really brought home Amazon’s dominance of the cloud services arena. There hasn’t been a better time to be developing web applications than now with the myriad of tools at our fingertips.
INTERESTED IN CLOUD SERVICES?
If you would like to talk to us about developing web applications using cloud services including A.I. drop us a line using the contact form or call on +44(0) 117 933 2595.