Phillip E. Trewhella
Full-stack software engineer and technical lead. Thirty years designing, building and running production systems end to end.
Available for consulting engagements · Open to remote or hybrid roles
Summary
AI has changed the landscape of software development, and with it the skills that matter most in building a successful product. Generating code is no longer the hard part — anyone with a subscription can produce a plausible function or a working page. But a product is a system of which the generated code is a small component. Designing, building, deploying, troubleshooting, maintaining and supporting such a system requires someone with a strong technical background.
Much of my career was spent writing code by hand. That part of the job has been largely commoditized — AI produces solid code for complex tasks orders of magnitude faster. Many in the industry perceive that as a threat. I see it as simply a shift in the way we work, and in the skills that are most valuable to a hiring organization. With the help of AI, a single experienced individual can now do the work of an entire team. An individual who can:
- Decide what to build, and what to build it on. Knowing the options, their trade-offs, and what is actually feasible — before committing to choices that last for years.
- Guide the AI that writes the code. Knowing what to ask for, and never shipping code that is not understood.
- Design the integration points. Where two systems meet is where products break: a payment event that has to land exactly once, a cache that has to expire when the data behind it changes, a webhook that must not acknowledge before the work is done.
- Own the whole lifecycle. Source control, environments, deployment, secrets and configuration, monitoring, and the maintenance that begins the day it ships.
- Minimize complexity. Every service, dependency and clever abstraction is something that has to be understood again later. Much of the discipline is in leaving things out.
Doing all of that well takes experience — years of building systems, watching them fail, and supporting them afterward. My most recent example is backtestresults.com: a live trading platform I designed, built, and operate alone, integrating a dozen third-party services across three environments, moving real money and placing real orders. I had been building it for a year before I brought AI into it. AI did not build this product; it helped me build it faster.
Experience
Designed, built, and launched a SaaS platform where traders review published trading strategies backed by transparent, verifiable backtest data, favorite the ones they follow, and subscribe to real-time signal delivery over configurable webhooks — acting on signals programmatically or receiving alerts in their own systems.
The analysis suite reports more than sixty performance metrics across eleven categories — risk-adjusted performance, advanced analytics, risk and volatility, statistical risk, efficiency and reliability, trade statistics, drawdown, trade analysis, periodic analytics, equity, and slippage — alongside an equity curve with drawdown and event markers, a calendar heatmap, and a navigable calendar that drills down to a graphical display of individual trades. Deeper study is supported by Monte Carlo risk and robustness analysis, walk-forward analysis, rolling analysis, trade distribution analysis, edge analysis (MAE/MFE), and trade efficiency analysis.
Built the core backtesting and strategy execution engine in Java/Spring Boot, connecting to Interactive Brokers for live trading and ingesting historical market data from Databento across equities, options, and futures through a Python data pipeline. Strategy results and analytics are stored on AWS S3 and served through CloudFront for fast, reliable delivery.
The consumer-facing web app is built in Next.js/React/TypeScript and deployed on Vercel, with Clerk handling authentication, Stripe powering tiered subscription billing with upgrade/downgrade flows, Neon serverless Postgres as the primary database, Sanity as the CMS behind the documentation and strategy overview content, and Loops managing automated email lifecycle campaigns. A separate internal dashboard (Vite/React) is where strategies are defined, backtests are configured and run, results are reviewed and published, and live execution is monitored. In total the system spans four codebases and roughly 80,000 lines of production code.
TriVirum, Inc.
Project Manager and Technical Lead (Contractor)
Led an offshore software development team responsible for building and maintaining TriVirum's cloud platform — the backend system that receives, processes, and distributes ECG data captured by the company's low-cost disposable screening device for early detection of atrial fibrillation. The device is prescribed to at-risk patients, worn for up to 15 days, and returned by mail for data extraction and analysis.
The platform was built heavily on AWS, leveraging Lambda, API Gateway, Cognito, RDS, Aurora, S3, Amplify, CloudWatch, CodeBuild, CodePipeline, and CloudFormation. Frontend applications were built in TypeScript with React and Redux, and healthcare interoperability was implemented with FHIR and SMART on FHIR. I was responsible for technical direction, sprint planning, and delivery across the full development lifecycle.
Murata Vios, Inc.
Interoperability Consultant
Served as the interoperability specialist for Vios's patient-worn vital signs monitoring system, used in acute-care settings. The system consists of a chest-worn sensor, a Bluetooth LE-connected Android tablet, and an on-premise server for data capture and hospital integration.
I was responsible for on-site installation and configuration, coordinating with hospital IT teams to connect the system to patient admissions and electronic health record systems. I also created the customer-facing technical documentation for the product's interoperability capabilities.
Sotera Digital Health, Inc.
Software Development Contractor
Assembled and led an offshore development team to enhance Sotera's remote alarming solution and build a new system management platform. The remote alarming software was developed for both iOS and Android and included a server component. I was responsible for team selection, technical direction, and delivery across all workstreams.
Sotera Digital Health, Inc.
Vice President, Systems Software and Interoperability
Joined as a Senior Software Developer and grew into the role of Vice President, Systems Software and Interoperability over an eight-year tenure. Sotera's product is a patient-worn wireless vital signs monitor for acute-care environments, continuously measuring heart rate, respiration, SpO2, blood pressure, temperature, ECG, arrhythmia detection, and fall detection — all transmitted in real time over the facility's wireless network.
I was responsible for building, deploying, and supporting the full software stack: the on-premise server that captures and stores patient data at full fidelity, the remote monitoring viewers installed at nursing stations, and the remote alarming system. I also owned interoperability — integrating with each customer's patient admissions, electronic health records, and third-party alarming systems using HL7 and related healthcare standards.
Earlier Career
Various Software Development Roles
Held a series of software development roles — both full-time and contract — across multiple industries and company stages, including two medical device startups. These early roles built the foundation for my expertise in systems software, healthcare integration, and leading small teams through ambiguous, fast-moving product development cycles. Happy to provide further details upon request.
Education
Purdue University
Bachelor of Science, Computer Science
Get in Touch
I take on consulting engagements, and I'm open to the right remote or hybrid role. Either way, the work I do best is the same: shipping a product end-to-end, standing up the services and environments around it, untangling a systems integration problem, or bringing a second opinion from someone who has been building software for three decades. Send a message or schedule a call — whichever suits you.