Taurus Terrell

Dynamic and results-driven full-stack engineer with 5 years of hands-on experience building evaluation systems and data collection pipelines for advanced AI applications. Expertise in designing automated test suites, analyzing large-scale user behavior data, and creating robust evaluation frameworks specifically for AI coding agents. Passionate about measuring how AI systems perform on real-world coding tasks, including code generation, repository migrations, and feature implementation.

Strong background in JavaScript/TypeScript ecosystems with proven ability to deliver scalable testing infrastructure and performance monitoring solutions.

Senior Full-Stack Engineer - AI Evaluation Systems at Anthropic (2024-06 – Present)

• Led the design and implementation of comprehensive evaluation frameworks for Claude AI coding assistants, developing over 500 automated test cases spanning JavaScript, TypeScript, Python, and Java.
• Architected and maintained high-throughput data collection pipelines processing millions of daily AI-user interactions, delivering actionable insights to improve model performance.
• Designed and deployed robust evaluation systems for real-world coding tasks, including large-scale JavaScript-to-TypeScript repository migrations, bug fixes, and complex feature implementations.
• Built automated testing infrastructure that improved evaluation accuracy for AI migration capabilities by 40%.
• Developed interactive dashboards and reporting tools used by research and engineering teams for data-driven decision making on AI model improvements.
• Collaborated with cross-functional teams of 20+ engineers and researchers to analyze AI agent behavior patterns and optimize system performance.

Full-Stack Software Engineer - AI Testing Infrastructure at OpenAI (2023-07 – 2024-05)

• Developed core testing infrastructure for ChatGPT and Codex-related development tools, creating automated evaluation pipelines focused on AI code generation quality.
• Built extensive test suites for multi-language code repositories with specialized focus on JavaScript/TypeScript AI agent performance assessment.
• Implemented scalable data collection systems that analyzed developer productivity metrics and AI tool adoption patterns across thousands of users.
• Created real-time monitoring systems tracking AI agent consistency across different programming languages and frameworks.
• Contributed to A/B testing platforms that evaluated new AI coding features with large developer cohorts.

Software Engineer - Evaluation Systems at IBM (2022-06 – 2023-06)

• Designed and implemented evaluation systems for IBM Watson AI-powered development tools, emphasizing automated testing and performance benchmarking.
• Developed data collection pipelines that processed hundreds of thousands of developer interactions monthly to assess AI impact on coding workflows.
• Created comprehensive test suites and migration evaluation tools for JavaScript to TypeScript conversions using AI agents.
• Built monitoring and analytics dashboards that provided insights into AI-assisted development efficiency and code quality improvements.

Software Engineering Intern at Salesforce (2021-05 – 2021-08)

• Contributed to the development of automated testing frameworks for Salesforce's AI-enhanced development platforms.
• Built data collection tools and performance monitoring systems for internal developer tools.
• Developed REST API test suites and evaluation dashboards used by engineering teams.

Software Engineering Intern at Adobe (2020-05 – 2020-08)

• Assisted in building early-stage evaluation tools and data pipelines for Adobe's AI research initiatives.
• Created automated test scripts and performance tracking systems supporting internal AI projects.

Master of Science in Software Engineering – Georgia Institute of Technology (2023-08 – 2025-05)

Bachelor of Science in Computer Science – University of Michigan (2019-08 – 2023-05)