About Me

I’m a senior machine learning specialist at AWS. I received my M.S. in mathematics with an emphasis in data science from Tarleton State University and previously was a senior data scientist and solutions architect at TechnipFMC.

For a more detailed view of my professional history, you can check out my resume. If you like aesthetic resume templates, get the template I used for free.

Open source projects and contributions

• openllmetry, added tracing support for Amazon SageMaker endpoints
• agent-evaluation, a generative AI-powered framework for testing virtual agents
• amazon-bedrock-samples: custom Amazon CloudWatch dashboard for Bedrock apps
• bedrock-vscode-playground, a VS Code extension that enables developers to explore and experiment with Large Language Models (LLMs) available in Amazon Bedrock
• vscode-zettel-archive, a VS Code extension that supports note-taking motivated by the Zettelkasten system
• sagemaker-scikit-learn-container, Amazon SageMaker’s pre-built Scikit-learn framework container where I fixed an issue regarding outdated boto3 and botocore versions which blocked usage of new AWS APIs from 2021 and 2022.
• sagemaker-python-sdk, the official Amazon SageMaker Python SDK where I fixed a model registry problem with some of SageMaker’s pre-built containers.
• aws-samples: sagemaker-feature-store-real-time-recommendations, a machine learning workshop given at re:Invent 2021 walking attendees though creating and deploying a real-time recommendation engine leveraging Amazon SageMaker Feature Store as a core component of the solution
• aws-samples: enhanced-pyspark-processor, a class override to enable local mode for SageMaker Processing with Spark
• aws-samples: ml-lineage-helper, a library around SageMaker ML Lineage Tracking extending ML Lineage to end-to-end ML lifecycles, including additional capabilities around Feature Store groups, queries, and other relevant artifacts
• amazon-sagemaker-examples, a contribution to AWS’s official amazon-sagemaker-examples repo in GitHub exploring the flexibility and practicality of Script Mode in SageMaker
• covid19, a repo containing code written to scrape COVID-19 data, do analyses, and generate content for the endcoronavirus.org website
• ml-model-lambda-microservice, deploy a scikit-learn model in AWS using Lambda and API Gateway
• ZenML, a custom data science and deep learning Python library
• stokepy, a stochastic models Python library
• atari pong ai, an AI that learned to play pong via reinforcement learning
• markov twain text generator, a Markov model implementation of a Mark Twain chatbot
• self-assembly, a particle physics simulation to gain insight about how particles self-assemble
• binning optimization of categorical levels, a statistical framework that decides which categorical levels should be kept before a model is built
• denoising dirty documents Kaggle competition, an OCR algorithm to clean up coffee-stained, sun-spotted, and wrinkled pages that have been digitized

Certifications

AWS

• AWS Certified Solutions Architect - Associate
• AWS Certified Machine Learning - Specialty

Stanford

• Divide and Conquer, Sorting and Searching, and Randomized Algorithms

Deep Learning Specialization, deeplearning.ai

• Neural Networks and Deep Learning
• Improving Deep Neural Networks
• Structuring Machine Learning Projects
• Convolutional Neural Networks

TensorFlow in Practice Specialization, deeplearning.ai

• Introduction to TensorFlow for Artificial Intelligence, Machine Learning, and Deep Learning
• Convolutional Neural Networks in TensorFlow
• Natural Language Processing in TensorFlow

External publications

AWS Machine Learning Blog

• Evaluate conversational AI agents with Amazon Bedrock
• Improve visibility into Amazon Bedrock usage and performance with Amazon CloudWatch
• Accelerate and improve recommender system training and predictions using Amazon SageMaker Feature Store
• Extend model lineage to include ML features using Amazon SageMaker Feature Store
• Bring your own model with Amazon SageMaker script mode

Conference talks & media appearances

re:Invent 2024

• Enhance ML workflows using SageMaker with MLflow and IBM integration [AIM378]: I was a speaker for a deep-dive session on managed MLflow on Amazon SageMaker

Toronto Summit 2024

• Accelerate development and delivery of FMs on Amazon SageMaker [AIM303]: I was a speaker for a session that provided FMOps best practices and tooling

New York Summit 2024

• Accelerate development and delivery of FMs on Amazon SageMaker [AIM308]: I was a speaker for a session that provided FMOps best practices and tooling

re:Invent 2023

• Deliver high-performance ML models faster with MLOps tools [AIM305]: I was a speaker for an advanced-level MLOps workshop covering MLOps tools on AWS that help you automate and standardize testing, managing, deploying, and monitoring hundreds to thousands of machine learning models.

re:Invent 2022

• Implementing MLOps to deliver high-performing ML models faster [AIM308]: I was a speaker for an advanced-level MLOps workshop covering MLOps tools on AWS that help you automate and standardize testing, managing, deploying, and monitoring hundreds to thousands of machine learning models.

MLOps World 2022

• Implementing MLOps Practices on AWS using Amazon SageMaker: I was a speaker for an advanced-level MLOps presentation and workshop covering machine learning operations with Amazon SageMaker.

DC Summit 2022

• AWS On Air: Using SageMaker Pipelines to build repeatable ML workflows
• Improving your machine learning in production with MLOps [AIM401]: I was the speaker for an advanced-level machine learning chalk talk where I walked attendees through core strategies and best practices for improving their machine learning operations.

re:Invent 2021

• Real-time recommendations using Amazon SageMaker Feature Store [AIM415]: I was the speaker (and content developer) for an expert-level machine learning workshop walking 100+ attendees through creating and deploying a real-time recommendation engine leveraging Amazon SageMaker Feature Store as a core component of the solution. For more details, see the workshop description:

Data scientists apply feature transformation on organizations’ raw data to generate aggregated features for their machine learning algorithms. These aggregated features are critical for real-time applications, such as fraud detection, real-time recommendations, and personalization. However, serving these types of features for real-time predictions in production poses a difficult problem. In this workshop, learn how to use Amazon SageMaker Feature Store to make personalized recommendations more responsive and improve viewer experience.

• Explore, analyze, and process data using Jupyter notebooks [AIM324]: I was the speaker (and content developer) for an advanced-level machine learning chalk talk where I walked attendees through newly-released SageMaker Studio features, specifically exploring and connecting to different data sources directly from the notebook and using that connection to analyze and process data, then train and deploy a model - all without leaving your notebook.