Nik Johnson Email & Phone Number

New York, Ny, Us

San Francisco, Ca, Us

Twitter API - StreamingWorked on the next version of Twitter's streaming APIs!

San Francisco, Ca, Us

Design and run a new program for testing new server SKUs within Twitter. Worked with dozens of teams to ensure hardware meets customer needs and is tested with customers as part of the entire server development timeline, from initial meetings with vendors to mass production.Worked with vendors and open source contributors to modify OpenBMC to meet Twitter's needs.Designed and implemented a service to automate running and logging of various hardware and OS level tests.

San Francisco, Ca, Us

Platform EngineeringConceived, obtained buy-in from management, and lead a project to build a new unified API platform to front all of Platform Engineering's platforms for use by all of Twitter Engineering. The project involved designing an API that could take a standard format and be translated into a multitude of requests to backend services to provision platform resources, such as compute quota, blob storage, KV databases, SQL databases, security resources, and more. The system included a job tree library and execution engine that was idempotent and resilient to failure.The resulting automation massively decreased the time engineers had to spend waiting for resources to be allocated, going from days when it was done by humans to seconds with the new system. It continues to be expanded and includes more resources to this day.Lead team in rethinking the internal resource ownership website into being the "Single Pane of Glass" for Platform Engineering resources.Worked to improve and expand the internal chargeback systems to improve knowledge among developers of how much their services actually cost to run.Mentored junior SWEs on navigating their careers and improving their skillsets.

San Francisco, Ca, Us

Provisioning EngineeringMaintained the services for storing the server and network gear directory, and for automated provisioning.Introduced a rollback feature to the provisioning service to undo failed actions, preventing partially initialized machines from landing in production.Envisioned, designed, and implemented a disk imaging feature for the provisioning stack, cutting down provisioning times from nearly two hours to on average 24 minutes. Root-caused and fixed several critical bugs in the directory service, massively improving reliability and resiliency of the service.

San Francisco, Ca, Us

Changed how Revenue SREs organized themselves from an embedded model to a strike team dedicated to Revenue. The new team was far more successful at working together to solve organizational wide issues, and deduplicating work across the individual teams we supported.Assisted in the design and build out of the full tweet search index, including performance testing, cluster management tools, and automation safeguards.Unified the software deploy strategy for the organizations I supported, greatly simplifying the steps involved and allowing engineers to move between teams easily.

Setup and maintain Linux and FreeBSD serversManage Zenoss cluster: Update operating systems and software and user and device managementWrite plugins for Zenoss to enhance monitoring of site-specific equipment

Austin, Tx, Us

Develop automated tests using Capybara to increase test coverage and reduce man hours needed for testingWrote automatic database query script to email statistics which saved hours per week spent manually gathering the information.