Summary of AWS blogs for the week of Monday, Apr 3

In the week of Mon Apr 3, 2023, AWS published 61 blog posts — here is an overview of what happened.

Topics Covered

Official Machine Learning Blog of Amazon Web Services

Using AWS Services to Improve Airport Baggage Handling Systems, Voice Chat Security, and More

Deploy a Predictive Maintenance Solution for Airport Baggage Handling Systems with Amazon Lookout for Equipment

Probably everyone has checked their baggage when flying, and waited anxiously for their bags to appear at the carousel. Successful and timely delivery of your bags depends on a massive infrastructure called the baggage handling system (BHS). This infrastructure is one of the key infrastructure components in any airport’s operations. Many airports are now looking to optimize availability and minimize downtime through predictive maintenance.

Moulham Zahabi from Matarat has co-written this post to show how to use Amazon Lookout for Equipment to deploy a predictive maintenance solution for airport baggage handling systems. Amazon Lookout for Equipment helps to detect and diagnose anomalies in equipment operations based on data collected from Internet of Things (IoT) sensors. It does this by combining metrics from all equipment signals and leveraging ML models, which learn from equipment signals to detect faults.

Amazon Lookout for Equipment also provides a fault tree-based approach to identify the root cause of equipment failures. It then provides recommendations for maintenance teams to address the root cause and resolve the issue, helping minimize downtime and improve availability.

Modulate Makes Voice Chat Safer While Reducing Infrastructure Costs by a Factor of 5 with Amazon EC2 G5g Instances

Carter Huffman, CTO and Co-founder at Modulate, has written this post to show how they are using Amazon EC2 G5g instances to reduce infrastructure costs by a factor of 5 while making voice chat safer. Modulate is a Boston-based startup on a mission to build richer, safer, more inclusive online gaming experiences for everyone.

The Modulate team is using Amazon EC2 G5g instances to reduce latency and increase throughput for their real-time voice chat system. By leveraging the AWS Nitro System, the Amazon EC2 G5g instances give Modulate the performance they need to support their large user base.

Modulate is also using Amazon Rekognition to detect real and live users and deter bad actors. It does this by matching the user’s face in a selfie captured by a device camera with a government-issued identity card photo or pre-established profile photo. Additionally, Modulate is using Amazon SageMaker to transform their retraining MLOps pipelines and build Streamlit apps.

Secure Your Amazon Kendra Indexes with the ACL using a JWT Shared Secret Key

Organizations have difficulty accessing critical business data dispersed among various content repositories in a streamlined and cohesive manner. Amazon Kendra is an enterprise search service that makes it easier for businesses to access this data. However, organizations must secure their Amazon Kendra indexes with an access control list (ACL) using a JWT shared secret key.

Using JWT tokens, Amazon Kendra can authenticate the identity of a user or an application. It then matches the user’s or application’s identity with an ACL associated with the index. If the identity matches an entry in the ACL, Amazon Kendra grants access to the index. Additionally, JWT tokens expire over time and must be renewed, so organizations can control access to their Amazon Kendra indexes.

How Games24x7 Transformed Their Retraining MLOps Pipelines with Amazon SageMaker

Hussain Jagirdar from Games24x7 co-wrote this post to explain how the company is using Amazon SageMaker to transform their retraining MLOps pipelines. Games24x7 is one of India’s most valuable multi-game platforms and entertains over 100 million gamers across various skill games.

Using Amazon SageMaker, Games24x7 was able to quickly launch the familiar RStudio integrated development environment (IDE) in a managed Amazon SageMaker environment. This allowed them to easily monitor training jobs and track model performance. Additionally, they created a secure and flexible MLOps platform using Amazon SageMaker Pipelines, which enabled them to run PySpark processing jobs within a pipeline.

Moreover, Games24x7 was able to use the Amazon SageMaker JumpStart feature to inpaint images with Stable Diffusion models. They also deployed large language models on AWS Inferentia2 using large model inference containers.

Cloud2 Can Help

Cloud2 provides both professional services and managed services to help customers maximize their AWS investments. Our Advanced AI writers can provide highly technical details and/or references to AWS documentation or code snippets in CloudFormation YAML or AWS API Calls using Typescript and AWS SDK for JavaScript v3.
We can also help you build, train, debug, deploy, and monitor your models and manage your machine learning (ML) workflows with Amazon SageMaker.
Additionally, we can assist you in securing your Amazon Kendra indexes with the ACL using a JWT shared secret key.

Contact us today to learn more about how Cloud2 can help you use AWS services to improve airport baggage handling systems, voice chat security, and more.

Read the full blog posts from AWS

Scroll to Top