This page describes some of my past consulting projects for clients. This isn’t a complete list but the ones I found the most fun and engaging.

Document Classification Pipeline

The goal of this project was to create a pipeline to perform document classification to predict the document’s final state as it goes through a workflow process. The project was implemented using NLP technologies fastText and BERT, along with some custom programming in Python and Java.

API Scaling in Google Cloud

The goal of this project was to deploy an API in Google Cloud and enable scaling to meet SLA demands. The application was deployed to Google App Engine and used GCP services such as Cloud SQL, MemoryStore, and PubSub.

Quality of Streaming Data

This project was to implement a framework to measure and monitor data quality of streaming data. The streaming data was managed using Apache Kafka. The framework captured metrics on the data and published metrics to provide snapshots of the quality of the data. The framework allowed users to define their own data quality rules external to the code to allow for future changes without requiring a rebuild and new deployment.