Brandon Coleman

This was a Windows Presentation Foundation (WPF) project that involved building an asset management system as an internal tool for the company. The system would allow field personnel to record and edit geographic assets, and the data would then be synced with our server.

Interesting additions to the software included integrations with differential (backpack) GPS units, Bluetooth GPS units, forward-looking infrared thermal cameras, onboard tablet cameras, synchronization with our remote server, automatic importing of spatially related data into input forms, and configurations through an attached database.

An ASP.NET MVC application to support a clinical trial by a pharmaceutical manufacturer. Clinical trial sites used the software to record patient information throughout the trial process. This particular trial included an imagery component of pathology slides that were generally large and were synchronized between client machines, our server, and Amazon S3 (AWS S3.) Images were processed with a machine learning algorithm to identify cellular characteristics concerning the trial. The process was run via a complex state machine, and the image processing was handled with a series of fault-tolerant jobs using Hangfire as a job queue.

The application allowed users to import all the information into the platform, upload images to process, and access information and imagery analysis results by medical professional contractors. At the end of the contract, the data was packaged up and shipped to the FDA for approval.

An RoR application that supports the business logic and workflows for the company's primary product offering. The application comprises an API and a series of Sidekiq jobs that manage client workflows and ledgers. The central data store of the application is Salesforce, which also serves as the app's front end.

I was a senior developer on the team, planning and leading several project initiatives. I migrated data models from Heroku Connect to Salesforce API, set up a series of jobs to rework client ledgers for client affordability and company revenue, planned efforts to convert potentially sensitive information to encrypted data, and planned work on large refactors from the initial MVP to a more stable codebase. Many problems during these tasks resulted from MVP-related shortcuts or acquired software/data that needed to be worked into the system.