GREY LATHROP

SOFTWARE & SYSTEMS ARCHITECT

You Want

A distinguished engineer with experience at every layer of the technology stack from the OS to the front-end. A leader, both technical and cultural, passionate and opinionated about best practices yet seasoned enough to recognize when Done beats Done Right. Take advantage of twenty+ years of professional experience:

• Designing, building, and deploying complex scalable cloud-native systems that perform well, degrade gracefully, and fail rarely, to help solve the complex problems that drive the business.
• Strengthening alignment across areas of the business by distilling complex technical concepts and systems into their essentials and communicating with non-technical and heavily technical stakeholers alike.
• Driving adoption of patterns and processes that manage the complexity of scaling the business both technologically and as a team.
• Leading by example and mentoring other engineers to help them grow into the technical virtuosos you need as the core of your engineering team.

I Have

• Broad and deep expertise in the Amazon Web Services (AWS) cloud computing platform and related services. Extensive, proven experience designing and delivering cloud-native applications which take advantage of the AWS ecosystem.
• Expertise in cloud cost optimization, creating cloud infrastructure and applications which scale economically as well as technologically.
• Expert understanding of software development methodologies and how they break down in real teams. Significant experience creating functional and pragmatic development processes for the team(s) I work with.
• Comprehensive high-level understanding of programming concepts and talent for applying those concepts to rapidly develop proficiency in unfamiliar environments.
• Expert polyglot programmer with significant practical development experience in several programming languages. Developed and deployed non-trivial applications and utility scripts in Rust, Scala, Python, Go, Java, Common Lisp, Ruby, Node.js, and Bash.
• Significant understanding of the underlying theory of computation and its application to solving more complex business problems through the use of meta-programming, domain-specific languages, etc.
• Expert-level understanding of scalable infrastructure design principles.
• Expert-level understanding of source-code management principles and systems, particularly git.
• Expert-level understanding of Linux/Unix administration and internals.
• Intermediate understanding of relational database systems, including scaling techniques.
• Experience providing technical leadership across several engineering departments (150-200 engineers).
• Experience scaling technical operations teams from a single engineer to over fifteen. Experience advising engineering teams in general on applying those scaling techniques to their teams.
• Experience managing a small team (2 direct reports).
• Developed complex software in Python; worked with common Python frameworks such as Tornado, Django, and SQLAlchemy.
• Developed complex realtime stream processing software in Java.
• Transitioned from full-time Operations to full-time Development and back. Gained a deep understanding of both sides of the "DevOps" divide.
• Built and lead an Operations team from the ground up; ensured Operations stayed highly integrated with Engineering as both teams grew.
• Designed, built and automated several multi-AZ infrastructures on top of Amazon Web Services.
• Participated in design and implementation of custom cluster management software, used to manage 1000s of servers across bi-coastal datacenters.
• Implemented Puppet modules to automate configuration of a broad range of services.
• Contributed to the maintenance and improvement of open-source tools.

I've Been

• Performed a hands-on operations uplift for an understaffed tier 0 service under deadline pressure in order to achieve critical business milestones.
• Led key portions of a fast-paced migration of MuleSoft Control Plane from self-managed Kubernetes infrastructure to AWS EKS under pressure of increasing disruptions in production. Guided 50+ teams with over 200 services in a zero (additional) downtime migration involving several 500+ node Kubernetes clusters.
• Led key portions of a multi-year migration of Mulesoft engineering from New Relic to Salesforce internal monitoring systems.
• Led architecture for the Production Engineering division of MuleSoft engineering. Drove adoption of engineering standards across an organization of 180+ engineers.
• Represented Production Engineering at organization-wide "Virtual Architecture Team".
• Identified and documented multiple opportunities for multi-million dollar cost savings.
• Architected and advocated adoption of blue/green deployment strategies which were not generally used in MuleSoft.
• Drove adoption of safe change practices and the standardization of MuleSoft engineering onto a single build and deployment pipeline.
• Mentored senior engineers in the design and development of an infrastructure event bus for a transition to event-driven architecture for production engineering systems.

• Designed, built, and deployed a distributed ad-hoc data processing system using Spark, Kubernetes, and Argo.
• Wrote a Terraform module for deploying the kubernetes autoscaler and node termination handlers for EKS.
• Participated in code reviews of significant changes across several large pieces of machine learning software.
• Improved performance of on-premises machine learning classifier data ingestion by several orders of magnitude.
• Developed several utility Terraform modules to codify best practices and provide reusable components.

• Designed and lead implementation of system(s) for deep operational observability of the Heroku platform. Drove conversations about observability across the broader Salesforce architecture organization.
• Lead architecture for Heroku's Production Engineering department. Help drive creation and evolution of the architecture organization inside of both Production Engineering and the broader Heroku org.
• Lead architecture for next-generation hybrid deployment model, focusing on cloud coherence and universal patterns/tooling.

• Introduced Embedded Platform Engineering model to Salesforce engineering teams.
• Read and absorbed Salesforce Architecture Strategy documentation, began work on applying to Salesforce DMP (formerly Krux) infrastructure.
• Mentored new Platform Engineering technical leads, helped transition team leadership of both Core and Embedded teams.
• Architected and executed transition of data collection architecture (hundreds of thousands of req./s.) to new-generation platform with zero downtime and zero data loss.
• Led lossless transition from legacy continuous integration systems to new CI/CD platform.
• Led architecture of standard libraries for metrics, logging, and command-line applications.
• Authored light-weight processes and policies to address frictions involved in scaling a team and integrating a team into an enterprise company.
• Mentored platform engineers in learning Python and Scala.

• Led initial integration efforts between Krux systems and Salesforce.
• Documented culture and standard operating procedures of Krux platform engineering for sharing with wider Salesforce engineering organization.

• Developed and drove adoption of standards for service lifecycle, handoff, and documentation across entire Platform Engineering team; standards subsequently adopted across multiple engineering teams.
• Developed and deployed automated build & release pipeline from verifying unit tests on pull requests through packaging, promotion, deployment to staging, and execution of integration & regression tests.
• Developed and deployed a distributed user matching webservice in Scala + Play! currently serving 13k QPS in production.
• Divided and scaled a large-scale Kafka cluster with minimal impact to dependent services. Improved capacity by 50% while simultaneously decreasing costs by 11%
• Re-architected an internally-developed Java application using open-source components (Logstash & Elasticsearch); completely eliminating the need to maintain a custom codebase while improving throughput by an order of magnitude, and improving uptime.
• Provided technical leadership across several engineering teams. Drive adoption of standard patterns and processes across teams.
• Established and lead the Embedded Platform Engineering team; chartered to be embedded domain experts in systems and operational concerns within the development team and to guide the development and delivery of scalable, reliable software following best patterns and practices. Similar to the Google SRE model, but independently developed and adapted for Krux.

• Designed, advocated, and implemented the reorganization of the DevOps team to scale the organization. Drove adoption of new patterns and processes to support the new model.
• Led the daily tactical operations of the DevOps team for a year. Assisted with vendor negotiation. Managed 2 direct reports; gave direction to a team of five engineers.
• Re-designed and re-wrote a core service that is central to the functioning of most other Krux services. Improved performance by at least 40% and as much as 60% for some cases. Improved stability, reliability, and maintainability.
• Built a standard library for python applications at Krux to enable developers to build applications the "Krux Way" without re-implementing standard components.
• Architected and lead the implementation of a real-time data processing system scalable to thousands of requests per second.
• Built a data collection service capable of ingesting 2000 data points per second per process using the Tornado framework and python.
• Built a websockets service for delivering real-time data to a thick web client application.
• Contributed bug-fixes to front-end angular.js code for Krux' primary web application.
• Developed internal administration tools for Krux' primary web application using the Django framework.
• Introduced Kanban process to improve team workflow and visibility.
• Evaluated several "Big Data" storage engines; chose and deployed DataStax enterprise to serve as the back-end data store for Krux' user data service.
• Mentored teammates in best practices for Python development and deployment.
• Mentored teammates in the use of git for version control.
• Tracked down persistent performance difficulties with Tornado-based services; implemented solutions enabling continued scaling of those services.
• Refactored puppet manifests to reflect best practices. Wrote Puppet modules for:
  • Managing cowbuilder environments for building packages.
  • Installing, configuring, and monitoring DataStax Enterprise.
  • Installing and managing Java versions.
  • Upgrading the linux kernel.
  • Managing persistent SSH tunnels.
  • Gathering system metrics via sysstat.
• Wrote python scripts to monitor a variety of services.
• Led standardization efforts for deployment scripts/processes.

• Wrote puppet manifests for deploying, configuring, and managing collectd for metric collection and monitoring.
• Wrote a python plugin for collectd to write metrics to statsd (due to native statsd plugin being unavailable for the version of collectd used at Lyft).

• Led a zero-downtime migration between server co-locations facilities.
• Managed infrastructure for production website, including MySQL master/slave replication.
• Researched and resolved several performance problems caused by poor query optimization.
• Advised developers on technology choices for full rewrite of site.
• Advised founders on recruitment of new developers.
• Researched hosted Drupal platform, assist in migration to hosted Drupal platform and hand-off to hosted solutions team.

• Introduced Kanban process to improve team workflow and visibility.
• Build automatic provisioning system with kickstart and Puppet.

• Assisted in the design of an "Operations API" which we will use to control our systems programatically for ad-hoc tasks like restarting services, running consumption tests, kicking off Puppet runs, etc.
• Built and open-sourced piro, a system for remotely controlling services in a plugabble and programmatic fashion.
• Created a system which fully records incoming requests and spools a rolling window of requests to disk in order to allow us to replay requests in the event of a failure where we would otherwise lose data.
• Contributed to development of user-facing API services.
• Responded to a malicious attack on infrastructure by a privileged user; prevented attack from destroying critical systems. Led recovery from attack and implementation of new security measures. Led investigation of attack which ended in prosecution of responsible party.
• Built Puppet manifests and bootstrap scripts to allow us to bootstrap instances to various roles without having to maintain AMIs.
• Built Operations team and acted as team lead: represented operational interests to the development team leads, established processes (scrum, kanban), set priorites in response to business needs, delegated tasks to team members.
• Migrated configuration management from Chef to Puppet.

• Actively worked to close gap between engineering and operations, foster transparency,
  and empower engineering to iterate rapidly while maintaining site stability.
• Participated in development of internal cluster mangagement software.
• Wrote command-line application to automatically allocate and provision servers,
  including options to specify minimum hardware requirements.
• Redesigned Digg infrastucture to take advantage of heavy automation & configuration management.
• Developed Puppet modules to automate deployment, configuration, and lifecycle management of key clusters.
• Developed FAI scripts which bootstrap systems from 'bare metal' to functioning Puppet clients.
• Assisted in migration from Subversion to Git as primary source code management system.
• Responsible for ensuring reliable operation of production, staging, and development systems.
• Perform code pushes and maintain change documentation.

• Built an infrastucture which ran smoothly for over two years with minimal intervention.
• Automated system configuration using Puppet.
• Firewall design, implementation, and maintenance.
• Create, deploy, and manage Xen virtualized servers.
• Patch managment and server maintenance.
• Proposed and implemented automated backup system.
• Deploy & maintain split-horizon DNS services.
• Replace Zenoss monitoring system with Nagios.
• Implement SNMP infrastructure.
• Deploy Zenoss monitoring system, notifications, and escalations.
• Created extensive documentation, including straightforward how-to procedures for common administrative tasks.
• Configure & maintain MySQL database systems.

• Created Twiki-based project management application.
• Identified key areas of network and process improvement, proposed solutions.
• Proposed, planned, and implemented single sign-on solution and corporate directory service.
• Provided desktop support for a heterogenous network of Mac OS X, Windows, and Linux desktops.
• Responsible for researching and procurement of best-of-breed equipment to implement the needs of supported organizations.

• 24x7 pager support for critical production systems.
• Planned and implemented migration from SiteScope to Nagios network monitoring, implemented custom service plugins, distributed & redundant architecture, performance metrics, and custom reporting interface.
• Planned and implemented OpenLDAP directory service; researched and proposed site-wide integration of applications with directory service.
• Administered heterogeneous network of several hundred nodes spread across several sites; maintained WAN links, VPN, remote administration between locations; maintained two data center locations with over 125 servers.
• Performed audit of poorly maintained RAID systems, implemented automated maintenance and reporting of RAID array performance and issues.
• Proposed, planned, and test-deployed Linux desktop solution for sales representatives to replace costly Windows desktops.
• Performed extensive documentation of server and network infrastructure.
• Researched, planned, and assisted deployment of Novell eDirectory and ZenWorks infrastructure.
• Assisted in migration from Checkpoint firewalls to OpenBSD firewalls.
• Assisted in software license compliance audit.
• Oversaw replacement of aging desktop-class hardware with server-class hardware.
• Researched and proposed migration plan from NT-style domain to Active Directory infrastructure.
• Researched and proposed deployment of enterprise XMPP chat service for internal messaging.
• Researched and proposed replacement of overloaded internal mail architecture with distributed, scalable mailing architecture.

• Planned and implemented migration from archaic Sendmail-based email system to a database-backed Postfix mail server, delegating specific tasks to system administration staff. Supervised programming of customized email administration software for new system.
• Replaced several outdated firewalls with OpenBSD firewalls. Wrote sophisticated firewall rule sets to filter and monitor network traffic. Created automatic log processing scripts to analyze firewall logs and produce daily reports.
• Created standardized server configuration procedures and documentation. Oversaw redeployment of servers to comply with standardized configuration.
• Created and administered network access/security policies and procedures.
• Maintained and administered heterogeneous intra-office network, including Windows 2000/XP and Mac OS X workstations, Unix file and web servers, wireless access points, and network printers.
• Planned migration of intra-office network to Windows 2003 Active Directory infrastructure.
• Remotely administered and maintained DNS, web, and database servers in several co-location facilities.
• Provided Tier II technical support.
• Oversaw system backup and network maintenance as well as server monitoring (Nagios) and IDS.
• Performed software license audit and inventory.
• Maintained relationships with hardware, software, and bandwidth vendors.
• Installed and administered DHCP server for automatic network configuration.
• Automated a variety of administrative tasks through shell scripts and custom programming.

• Administered web, DNS, and database servers running FreeBSD, NT 4, and Windows 2000 Server.
• Administered MySQL and MS-SQL 2000 databases.
• Improved uptimes by as much as 29%.
• Created, implemented, and enforced security policies on network, server, and individual workstation levels.
• Managed server migrations to/from varios operating systems.
• Planned and installed internal networks: CAT5 cabling, network ports, high-bandwidth switches, wireless access points, routers, firewall, and VPN.
• Secured wireless network segment against unauthorized access and packet monitoring.
• Proposed, planned, and implemented conversion of aging collection of 486 PCs into Citrix Metaframe thin-client terminals.