Nick Willard

• AI Engineering System: Built a team-wide AI coding workflow using Claude Code custom skills and automated weekly coverage-report pipelines, scaling sprint velocity from 12–15 to 45–50 points per sprint (roughly 3x) and cutting PR cycle time from multi-hour queues to under 30 minutes.
• Billing Platform Foundation: Architected the production billing database from schema design through cluster provisioning, covering partition management, Flyway migrations, and security group configuration. Authored failure-mode playbooks that enabled self-service incident response across the team.
• Observability & Release Velocity: Implemented distributed tracing across all billing services in Splunk, unblocking alert configuration for the full billing system. Automated Jenkins and Docker pipelines that shifted releases from bi-weekly to daily continuous production deployments, cutting manual toil by roughly 45%.
• Resilience Engineering: Refactored key services with circuit-breaker patterns and live dependency tests, converting a brittle batch boundary into a fault-tolerant, observable service. Standardized team-wide PR templates and pre-commit hooks to harden the delivery process.

• Architected the decomposition of a high-availability 3D Secure (3DS) legacy monolith into modular Spring Boot microservices supporting millions of global authentication requests, increasing deployment independence and maintainability by 40%.
• Designed event-driven architecture with the Axon Framework, improving system responsiveness and throughput by 30%.
• Led Oracle → PostgreSQL migration reducing licensing costs by 50%+; integrated Redis caching to accelerate high-frequency reads by 20–35%.
• Enforced 90%+ test coverage across microservices with JUnit/Mockito, reducing production defects by ~25%.

• Global Data Pipelines: Designed and operated multi-region ETL pipelines on Apache NiFi, routing terabytes of business-critical data across global environments. Automated Shell deployments that cut migration-related incidents by 40%.

• Integration Tooling: Connected Salesforce, Jira, and Zendesk APIs to generate automated cross-system reports, eliminating recurring manual work for 50+ business and ops stakeholders.
• Developer Automation: Built PowerShell and Chocolatey automation that cut workstation onboarding from 4+ hours to under 30 minutes, reducing recurring maintenance toil by 60%. Modernized the frontend CI pipeline to Gulp 4+, achieving 35% faster transpilation and roughly 25% fewer build failures.
• Observability: Implemented AppOptics APM across a global event-management platform (stadiums, convention centers), reducing mean time to detect (MTTD) by 40%.

• System Architecture: Built a fully serverless AWS coaching pipeline (EventBridge, Lambda, Bedrock, SES) that ingests live Strava activity streams and Coros biometric data including HRV, sleep score, and resting HR. Computes TRIMP-based training load metrics (CTL/ATL/TSB) and generates personalized daily HTML briefings via Claude.
• Infrastructure & Cost Engineering: Provisioned the full stack with Terraform via GitHub Actions OIDC with no long-lived credentials stored in CI. Cognito-authenticated REST APIs serve both a React SPA and a React Native mobile app from a single backend. Full system cost is roughly $1/month per active user.

“All numeric reasoning and anomaly detection runs in Python before the prompt is built, so the model only ever sees validated, typed inputs. Compute first, generate second.”