Lloyd Cotten

Overview

• Oversaw the implementation of Kubernetes in Amazon Web Services (AWS) for a microservice-based commercial real estate platform with 8 million properties and 1000s of users while optimizing application and AWS cloud infrastructure saving 5-10% of costs each year.
• Achieved 90% infrastructure as code coverage in Terraform, including AWS, Cloudflare, and UltraDNS infrastructure.
• Continuously improved security posture with more encryption, TLS at every level, WAF, network firewalls, and vulnerability monitoring.
• Managed both SQL and NoSQL databases, optimizing queries and indexes, designing disaster recovery strategies, performing version upgrades, and right sizing.
• Designed and maintained Gitlab CI/CD and executed a migration to GitHub and CircleCI.
• Mentored several team members and managed a small team of DevOps engineers.

Security

• Rolled out MFA, SAML, and OpenID Connect to enterprise authentication provider (Okta) for single sign-on (SSO).
• Performed access audits and applied security best practices such as least privilege, encryption and credential rotation for 100s of secrets using role-based authentication control (RBAC) implementations in Kubernetes, AWS IAM, and AWS KMS.
• Designed segmented network using Gateway Load Balancers (GWLB) and traffic patterns inspected by Palo Alto firewall appliance.
• Monitored application security with Snyk for SAST and Netsparker for DAST.
• Utilized Qualys for vulnerability management and Crowdstrike Falcon for 75% reduction in known vulnerabilities and zero exploits.
• Implemented cross-availabilty zone (AZ), multi-region, and multi-cloud disaster recovery (DR) strategies.

CI/CD

• Maintained self-hosted Gitlab instance for code, issue tracking, and CI/CD.
• Configured and managed CI/CD pipelines for both application code and laC (Terraform and Ansible) using Gitlab, GitHub, and CircleCI.
• Templated CI/CD YAML snippets for re-use across dozens of microservice repos to easily control deployments to non-production and production environments.

Observability

• Rolled out self-hosted Grafana to monitor, visualize and alert based on system metrics and status using metrics and logs from InfluxDB, CloudWatch, Elasticsearch, Prometheus, VictoriaMetrics, and other data sources.
• Integrated Grafana alerting with endpoints including PagerDuty, Squadcast, Slack, and Microsoft Teams.
• Participated in on-call rotations and managed scheduling with PagerDuty, easily sur

Overview

• Implemented several significant time-saving enhancements, saving meteorologists up to 2 hours of tedious data entry per day, and making new model data available 40 minutes faster every 6 hours.
• Created several new applications, with most apps achieving 100% code coverage in unit testing.
• Advanced the concept and implementation of stateless, microservice, and event-driven architectures for improved scaling, self healing, and reliability.
• Introduced Docker/containerization for improved scaling, ease of deployment and configuration, and re-use.
• Provided training, mentorship, and leadership for a team of college graduates developing new weather forecasting software.

Weather Data Publishing Software

• Maintained and supported legacy C# .NET desktop application used by meteorologists to produce thousands of weather forecasts daily.
• Parallelized forecast initialization to save forecasters 20-30 minutes each shift.
• Added new features such as forecast interpolation, snow ratio calculations, wave fetch length algorithms, and forecast confidence.
• Participated as architect for a new web based version, utilizing modern tech and best practices: Node.js, React, unit testing, and CI/CD.
• Developed and maintained multi-tier C# software consisting of a front-end web app and clustered back-end service created PDFs, HTML content, email, charts, CSV, and XML files that presented weather data to clients in templated formats, delivering 1000s of documents daily on schedule.

Model and Observation Processing

• Developed a Node.js application that parallelized downloading and processing of weather models, decreasing the processing time for GBs of model data to 20 minutes from 1 hour.
• Developed highly available, scalable, and modular set of Node.js microservices for retrieving, parsing, and transferring weather observations to various database, file, or email destinations.
• Used SQS to queue observation messages between the microservices.
• Fully horizontally scalable - new instances could spawn to meet queue message backlogs.
• Deployed to AWS using CloudFormation, EC2 Auto Scaling, ECS (Docker), EFS, and Route53.
• Followed a test-driven development method, achieving 100% unit test code coverage.

• Architected and developed Infotech’s first software product line of 5 new SharePoint web parts and components.
• Created automated build system that compiled binaries, documentation, and installer using NAnt, Microsoft Cabinet SDK, and Inno Setup.
• Evangalized and became an early adopter of a development process based on Agile/Scrum methodologies.
• Lead quality assurance initiatives, defect tracking, and daily build and deployment processes.

• Created campaign management software for voter outreach efforts, contributing to successful electoral campaigns.
• Delivered web site development services and developed an online reservation system for a hospitality client.
• Wrote a custom policy management application for an insurance agency.