Darryl Green Jr

Since joining The Washington Post, I've worked at the intersection of engineering and data privacy building the infrastructure that keeps the site compliant with an evolving set of global privacy regulations, while trying to make compliance itself less of a manual burden on other engineers.

One of the core projects was building a RESTful privacy API that consolidated privacy protocols across dozens of previously siloed projects. Before this, updating a privacy requirement meant touching each project individually; afterward, a single update could propagate sitewide, cutting future update requirements by an estimated 90%+.

I also led a cross-functional Agile team through sunsetting tracking cookies for EU users to bring nearly all Washington Post surfaces into GDPR compliance, closing gaps identified during legal's compliance review. This meant coordinating between engineering and legal throughout, and using AI-assisted development to manage refactoring efficiently across 15+ repositories. In a related effort, I rebuilt the site's cookie consent management interface to proactively close gaps against a patchwork of regulations — not just GDPR, but state-level laws like California's CCPA and Virginia's CDPA.

Beyond direct engineering work, I partnered closely with the legal team to identify and remediate compliance gaps across Post systems work that, per legal and leadership's own risk assessment, helped avert an estimated $3-10M in potential litigation exposure. I also authored Engineering Resource Documents and ran post-mortem analyses on completed projects, work that led to standardized testing and release protocols now used across the team, streamlining onboarding for new engineers.

On the pure engineering side, I built a shared HTML Web Component to streamline footer updates sitewide eliminating a recurring, redundant task that had previously required multiple developers to maintain. Throughout, I've leaned on AI-assisted development tools like GitHub Copilot to accelerate delivery across the codebase.

Tech Stack: React.js, Next.js, TypeScript, Playwright, Jest, HTML Web Components, Jenkins CI/CD, AWS, GitHub Copilot, Claude AI

At Michigan Aerospace, I worked across several distinct projects rather than a single product line, which gave me early exposure to a wide range of problems.

I architected a library of modular Vue.js UI components used across multiple application modules, which significantly cut development time for new features and brought consistency to a previously fragmented user experience. Separately, I inherited a Profile page redesign that an external contracting firm had left unfinished and, in some places, broken — malformed fields, failing validation — and took it through to completion ahead of the company's subscription model launch, improving data accuracy and reliability in the process.

I also built automated unit and integration testing suites for the Springmatter application using Jupyter Notebook and Python Selenium, which let the team catch regressions earlier in the development cycle rather than after they'd caused production delays. On a lighter but still valuable front, I integrated weekly automated testing suites simulating real user interaction ahead of production deployments, shaving 30 minutes off testing execution time through automation.

Alongside this work, I contributed to two distinct, concurrent initiatives. The first was a company-wide contract with Michigan State University researchers to build a camera-deployment application for identifying endangered species from field cameras placed at sites across the state — a project that grew significantly in scope over its lifetime. As part of the engineering team on this contract, I built a custom Python web scraping tool to compile the image datasets the project needed, and developed scoring logic that assigned species-match confidence likelihoods to field-captured images, feeding structured data into an LLM the team was training to recognize endangered species across camera deployments spanning dozens of locations statewide.

Separately, I engineered a Python-based tool for a Michigan Department of Natural Resources water turbidity initiative, converting deployed field images into video to significantly speed up analysis and improve data accuracy for the research team using it.

Tech Stack: Vue.js, JavaScript, Python, Selenium, Jupyter Notebook, HTML/CSS

A short but hands-on contract role at Ford Motor Credit, working within an existing Angular front-end codebase. I paired and mob-programmed daily with the team to debug and extend the application, building solutions against user-story acceptance criteria for functionality, scalability, and performance. I also performed weekly software testing and code reviews ahead of production launches.

The most concrete outcome of this role was engineering and deploying client-server A/B testing using Angular and Spring Boot — collecting real customer data and using it to drive targeted enhancements that improved conversion rates by 15%.

Tech Stack: Angular, TypeScript, Java, Spring Boot, JavaScript, HTML/CSS, Jenkins CI/CD