My Writing

Automate broadcast video monitoring using machine learning on AWS

In this post on the AWS Media Blog, I share code and lessons learned on building a processing pipeline for livestream video and applying machine learning to automate monitoring of broadcast channels.

Parallelizing across multiple CPU/GPUs to speed up deep learning inference at the edge

In this post on the AWS Machine Learning Blog, I discuss two approaches for parallelizing deep learning inference on a single device at the edge across multiple CPU/GPUs, providing example code snippets and performance benchmark results.

Preparing training data for object detection models on Amazon SageMaker

In the first post of a 3-part series on the AWS IoT Blog, I share best practices on using Amazon SageMaker GroundTruth to prepare training data for building object detection models on SageMaker.

Training custom object detection models on Amazon SageMaker using the built-in SSD algorithm

In the second post of a 3-part series on the AWS IoT Blog, I share best practices on using Amazon SageMaker GroundTruth to prepare training data for building object detection models on SageMaker.

Deploying trained SageMaker object detection models to the edge

In the second post of a 3-part series on the AWS IoT Blog, I describe how to deploy object detection models trained on Amazon SageMaker to Internet of Things edge devices for inference using AWS Greengrass

Voice applications in clinical research powered by AI on AWS – Part 1: Architecture and design considerations

In this post on the AWS for Industries Blog, I discuss Voice UI design and security/compliance considerations when building a voice assistant for patient outcome reporting in clicnial research.

Voice enabling participant diary collection powered by AI on AWS – Part 2

In the second post of a 2-part series, I walk through the technical details of building a patient diary sample application.

Discovering and indexing podcast episodes using Amazon Transcribe and Amazon Comprehend

In this post on the AWS Machine Learning Blog, I walk through a serverless podcast indexing application using AWS machine learning (ML) application services.