Adam Farrell — Resume

Cloud governance and platform engineering across Nike’s AWS accounts.

• Built the development roadmap for governed emergency access to AWS accounts, covering approval paths, incident integration, audit visibility and compliance reporting.
• Designed and operationalized the AWS account approval workflow, focused on governance, cost management, security controls and consolidation, with more visibility for engineering leadership.
• Co-created a managed AWS account strategy that uses Kubernetes control planes and Crossplane to run a multi-tenant environment.
• Co-host the weekly cloud governance and platform engineering community of practice.

Promoted to Principal, focused on retail in-store checkout and customer service.

• Cut P50 network latency in Nike Korea retail stores by 40% by optimizing network topology, working with corporate security and networking teams to establish a trusted origin.
• Implemented the AWS controls behind K-ISMS certification, affecting more than 3,000 users, with legal and corporate security teams.
• Helped launch CS Agent in the Nike App, replacing a legacy rule-based chatbot with an AI agent that understands intent and responds in real time. Athletes can track orders and manage returns, and the agent hands off to Consumer Services with a chat summary when they need more help.
• Reduced spam chat contacts to call centers by 75% and refund orders by 50% with a low-code solution.
• Wrote the transform step of an ETL process that migrated Korea customer service cases from Zendesk into Salesforce.
• Helped a vendor team with an AWS Glue migration pipeline for China user data, taking its total runtime from 28 hours to 34 minutes.
• Identified 30% AWS cost savings on a tiger team targeting organizational efficiency.

Promoted to Lead on the Sport Activity and Targeted Marketing Platform teams. In August 2018, moved to a new backend API team building Kotlin services for Nike’s mobile commerce apps.

• Took ownership of the Nike App home page API: up to 60% lower global latency, 4% more revenue through the channel, high-severity incidents down from one a month to none for two years, and dynamic content through Adobe Target.
• Built the order history service for nike.com and the mobile apps, consolidating several downstream APIs, moving logic out of the clients and cutting traffic to the shipping APIs by 30%.
• Got AWS Global Accelerator approved for production through thorough testing and documentation.
• Integrated SignalFx metrics and tracing into Nike’s Node.js GraphQL implementation.
• Spoke at AWS re:Invent 2018 to more than 500 attendees on how the Sport Activity team used DynamoDB to reach massive scale at low cost: DAT320, “Becoming a Nimble Giant: How Amazon DynamoDB Serves Nike at Scale”.
• Wrote “Moving Faster with AWS by Creating an Event Stream Database” for the Nike Engineering blog, about one of the entry points to the Targeted Marketing Platform.

Joined to work on the activity platform behind Nike Run Club and Nike Training Club: event-driven with CQRS and eventually consistent data stores, on AWS (SNS, SQS, EC2, CloudFormation), written in Scala and Java.

• Migrated activity and user time zone data from unmanaged Cassandra clusters to Aurora and DynamoDB, cutting DevOps work from days to minutes and AWS costs by 98%.
• Added ElastiCache (Redis) to the platform for granular data locking and fast payload retrieval by downstream systems.
• Wrote the orchestration service for DPA and GDPR requests across the activity platform, with easy auditing and replay on errors.
• Hosted and helped organize two record-setting Engineering Forums, each with more than 100 people in person and 50 online. The first brought in AWS Aurora experts through Nike’s AWS account manager; the second had Nike engineers speak about their work with containers.

Built Java and Spring web applications for GameHouse.com. Promoted to Senior in April 2013 for code quality, design decisions and mentoring, and joined the Business Intelligence team in July 2014.

• Designed a new shopping cart built for simplicity and easy extension, bringing cart processing from 3 seconds to 150 ms. Later moved it onto a new data services platform and added credit card processing and tax calculation.
• Key member of the project that moved GameHouse.com and its data off the RealNetworks platform onto GameHouse’s own, including the scripts and plan for migrating user accounts, purchase history and subscriptions from Oracle to SQL Server.
• Led the migration of the in-house marketing email system to an outside provider: schema, daily data updates and fixes to legacy templates.
• Built an SEO feature that highlighted and linked phrases in dynamic content, processing 10,000 content blocks against a 10,000-phrase dictionary in under a second.
• Made the platform changes for the company-wide PCI compliance effort.
• Improved production stability with feature flags the team could switch without touching a front-end server.
• Helped build a rewards program and a social toolbar to make GameHouse.com more socially connected.
• On the BI team, built solutions with PostgreSQL, Hive, Hadoop and Pentaho, and added Hadoop cluster monitoring that led to 25% better performance through configuration and uptime changes.
• Automated the game sales selection workflow, from a full day of manual work to half an hour.

Promoted from SDET to QA Architect after one year, then moved to the mobile team to work on the interaction between devices and servers.

• Implemented the instant messaging protocol and mail backend for the new MySpace iPhone app.
• Maintained and redesigned mobile notification services: SMS, BlackBerry and Apple Push Notification Service.
• Maintained and updated more than 200 iLike.com artist iPhone apps.
• Analyzed server-side performance of the site redesign, improving page throughput two to three times depending on the type of traffic.
• Designed a two-layer C# automation framework covering every page on the site, and a runner that distributed its tests across many machines and stored results in SQL for analysis.
• Gave technical guidance to a yearlong contractor project in India that expanded automation coverage across the site.
• Improved the international experience with a pseudo-localization culture and reverse key lookup for fixing translations, and a resource scanner that found more than 2,000 valid issues.
• Tested new features for a team of twelve developers, and wrote build scripts that let QA and product managers try UI flows before the backend was built.

One-year contract through Excel Data Corporation on the Windows International team, finding globalization and localization issues.

• Promoted from SDET I to SDET II after nine months, taking on duties from an employee and a departing contractor.
• Built language-neutral UI automation with MAUI for Active Directory tools: LDP, Domains and Trusts, Sites and Services, the ADAM setup wizard and AD Schema Analyzer.
• Triaged UI issues in international builds of Windows Vista, cutting handling time per issue from two minutes to one second with a fix rate above 90%.
• Ran the team test lab for Longhorn Server, setting up regression and debugging environments.