I joined Maplewell with its initial software development team of four. We created an edge appliance with supporting cloud infrastructure from scratch. My focus was on infrastructure for developer and power engineer enablement, and on the OS and runtime environment of the appliance.- Created a CI/CD process that was familiar to developers with a SaaS background, despite our reliance on edge appliances. It built, tested, and deployed each change to our entire fleet in around 10 minutes.- Enabled application metrics and log collection early on, leading to quick issue discovery and resolution.- Helped design and implement automatic package and tooling updates. This kept our edge environment and applications up to date, typically without requiring human intervention.- Configured zero-trust network to give local and remote engineers secure access to in-office hardware, to cloud resources, and to other engineers' local software instances for collaboration.As the business sought the right product shape, we maintained flexibility with a microservices architecture. Our applications could be deployed in diverse environments but expect similar infrastructure to be available to them.- Tailored infrastructure for local development, cloud, and edge environments to be appropriate to available hardware while presenting essential features and interfaces to applications.- Maintained set of application templates for developers. The team used them to experiment while maintaining clear separation of concerns. More than 20 python applications were created by this small team in less than a year.- Built a data pipeline to reliably stream time-series sensor data and events between edge and cloud, despite the intermittent network connections available to our edge appliances.- Designed and implemented secrets tooling that kept edge appliance secrets unique and secure while offering a similar and easy-to-use experience for local development.

I was charged with leading a CI/CD transformation early in my time at Automox. I joined shortly after the engineering organization had grown from one to multiple development teams, which had presented challenges in testing and releasing.- Collaborated with stakeholders at all levels to design the pipeline and process.- Implemented the changes across build tool chain, application configurations, and deployment environments.- Taught many groups in the organization how to use the pipeline, what to expect, and how to handle unusual situations.- The release cadence accelerated from 1 change out of 1 team per week, to multiple changes made by multiple teams concurrently almost every day.I was a technical leader on the Site Reliability Engineering (SRE) team. We were responsible for infrastructure supporting the Automox application and its development, from servers to storage to networks to load balancers to build pipelines to observability.- Worked closely with application development teams to understand their needs and plan supporting efforts alongside other work.- Transformed live environments to infrastructure as code. This led to improved team knowledge sharing and greater similarity between environments.- On-call first responder, responsible for rapidly troubleshooting and mitigating issues while communicating appropriately with many parts of the business.- Followed up on incidents by sharing lessons learned, improving documentation and dashboards, and adding alerting.When the SRE team’s manager resigned, I managed the team for several months. Once I had helped hire and onboard the new manager, I moved back to a more technical role.- Supported team of 5 to 6 in performance reviews and ongoing development discussions.- Designed hands-on interview session that provided valuable insight and received frequent positive feedback from candidates.- Instrumental in interviewing and onboarding 15 new SREs in 2021 onto our team of 5.

I developed core storage code for our distributed storage system, which was a set of highly multithreaded Linux applications running across a cluster of servers with many solid state drives.- Refined a CPU-intensive developer tool that performed low-level consistency checks into a tunable feature with low performance impact.- Optimized garbage collection system and added tuning parameters so that it could maintain effectiveness in more diverse customer scenarios.In addition to standard software development activities, I identified frictions in our developer experience and worked to resolve them.- Spearheaded hardening effort that reduced the failure rate of key unit test suite from 14% to 2%.- Created an internal developer tool which decreased unit test suite runtime by 70% (20 minutes down to 6) without requiring additional hardware.- Built a continuous test system for a library used by our product installation tools and developer tools. This dramatically reduced the number of defects slipping past release.- Engineered CI/CD pipeline for building and testing important OS functionality across many potential hardware variations, allowing us to confidently keep appliance packages updated.- Led the working group that coordinated and prioritized changes for developer tools across the teams who used them.

I maintained and enhanced applications on several different systems. I worked on low-level commanding and modeling of satellites, as well as the applications spacecraft operators used for generating optimal tasking schedules.- Maintained and enhanced legacy codebases with care to ensure the operational success of the spacecraft system.- Developed new features for a critical application on schedule, while also simplifying it enough to decrease total code size by 40% (18k lines of code to 11k).- My software designs were briefed to high-level government customers as examples of our high quality development process.- Represented technical team as a liaison to our customers during significant test events and deployments.- I fixed bugs and added features to code older than I was at the time, and code I wrote more than a decade ago is almost certainly still being used for spacecraft operations.