About Me
I'm a Staff Software Engineer working on backend and distributed systems. At Fivetran I work on the MySQL and Oracle database connectors that replicate customer data, mostly making them faster and more reliable. I like digging into hard-to-diagnose problems, whether it's change data capture, enormous schemas, or JVM memory.
Working remotely since 2020, and with teams spread around the world since 2012, I'm comfortable working asynchronously through documentation and written communication.
How I Work
Features start with a design doc that states the expected outcome, such as a UX improvement or a target metric, and a benchmark of current behavior taken before work begins. Regressions in performance or user experience need a good reason.
Decisions are documented in one central place and signed off by the people involved, to avoid scope creep and rework.
I mentor through one-on-ones and code reviews, with occasional pair programming, and share knowledge asynchronously in writing. I help unblock teammates even when the problem is outside my scope.
Decisions are made with the customer and the product in mind. Understanding customer needs leads to the right outcome.
Experience
Fivetran
Staff Software Engineer • September 2023 to Now
Technical lead for the MySQL connector, with a focus on reliability and performance. Also led several broader projects on the Oracle Binary Log Reader connector, and contribute to cross-team database connector architecture.
Solved out-of-memory failures in source metadata fetching for large schemas. It now fetches only the minimum data required for a customer's sync and keeps a consistent view of metadata between syncs. This reduces load on customer databases and eliminates schema drift. Metadata gathering for the largest customers dropped from about 20 minutes to under 5, unblocking a customer with roughly 480,000 source tables in a single connector.
Reduced incidents through better error handling and messaging, support playbooks, and JVM heap reductions.
Senior Software Engineer • March 2022 to September 2023
Led the connector reliability project, including connection-handling fixes that improved uptime, and a MySQL and MariaDB driver upgrade.
Led Elasticsearch and OpenSearch connector work to beta, and ran a performance-focused project that started with benchmarking to set a baseline.
Prototyped a feature-flagged driver upgrade with automatic fallback for another team, and mentored engineers and an intern.
Rocketmiles
Staff Software Engineer • July 2021 to March 2022
Key contributor to the platform unification project, migrating the original Rocketmiles.com site onto the shared white-label platform with no downtime.
Introduced Kotlin and Spring for new back-end modules to improve type safety and testability.
Managed platform resources to keep white-label sites stable through viral, promotion-driven traffic spikes.
Owned the pricing and hotel inventory platform fed by parent-company supply data, and launched non-hotel accommodations. Mentored the team building the greenfield cars booking platform.
Senior Software Engineer • September 2019 to June 2021
Back-end engineer on Rocketmiles.com who also contributed to the React and Angular front end. Worked fully remote from 2020.
Fixed JVM memory leaks in the Gradle back end using heap dump analysis and memory profiling.
CarrierDirect
Senior Software Engineer • September 2018 to September 2019
Tech lead for a truck-load shipment management platform (Java, Kotlin, Spring Boot, React, Python Lambda), including the move from CosmosDB to PostgreSQL.
Built the Kubernetes microservice template and shared authentication service reused across company projects, introduced DevOps practices, and mentored engineers on my team and others.
Label Insight
Software Engineer • September 2016 to September 2018
Built an event-driven content system (Lambda, Kinesis, DynamoDB streams, Elasticsearch) that publishes SmartLabel page updates the moment a customer approves a product, plus a global ingredient dictionary for bulk changes across a customer's whole portfolio. Customers can now edit any part of their product data directly, and verification went from weeks per phase to instant. This cut the internal data team's manual workload by more than half.
Created a centralized logging pipeline (Kinesis, Firehose, Elasticsearch), moved services to Docker on ECS, and codified infrastructure with Terraform and Serverless Framework.
Column Technologies
Software Developer • September 2012 to September 2016
Resolved major production incidents for enterprise customers (including one on-site visit) and built the Investigative product's similarity-matching engine on the BMC Remedy platform.
Re-architected modules for a 75%+ compute efficiency gain, and worked daily with a team spread across North America, India, and Australia and New Zealand.
Skills
Languages
Distributed Systems & Data
Cloud & Infrastructure
Frameworks & Practices
Education
Knox College
B.A. in Physics with Computer Science and Mathematics minors • 2012