Michael Knowles

summary

Backend ownership across the API (Go), third-party apps platform (NestJS/TypeScript), and AWS infrastructure (Terraform); 1,000+ merged PRs

highlights

• Architected and shipped Glue's integrations with the Model Context Protocol (MCP), building both server and client: an MCP server, an OAuth 2.1 authorization server and agent tool surface making Glue addressable by external agents like Claude and ChatGPT; and an MCP client, connecting Glue out to third-party MCP servers handling ~40,000 tool calls/day
• Owned the third-party integrations platform end-to-end, shipping 12+ production integrations (GitHub, Zoom, Google Workspace, Linear, Notion, Jira) driving ~6,000 app installs
• Drove the engineering execution of a company-wide infrastructure cost reduction: audited services, removed redundant vendor dependencies, and re-platformed the legacy RAG pipeline
• Implemented production secrets management (SOPS) and security response, including same-day remediation of customer-escalated security issues
• Re-architected the notification pipeline onto an asynchronous background-task queue, improving request latency and delivery reliability, and enhancing the messaging platform's data model
• Rebuilt platform observability with distributed tracing and structured logging, used to diagnose and resolve production incidents
• Reviewed 1,400+ pull requests as a top-2 reviewer on every core repository while serving as the escalation point for cross-team integration and production incidents
• Drove backend performance optimizations (indexing, N+1 elimination, query-planner tuning) that cut p95 query latency from 96ms to 10ms on a platform serving 42,000+ users and 33M+ messages

highlights

• Promoted from Senior Software Engineer, January 2023
• Led a team of engineers developing a SaaS platform in Go, Node, React, and Google Cloud. The platform analyzes $24 billion in sales annually, and sends 2 billion messages annually; it peaked at over 2,100 requests per second and 134,637 messages sent per second
• Architected a monolith into microservices, establishing unit/integration tests along with feature flags to ensure behavior parity
• Led the database migration to Google Bigtable and Postgres to reduce latency, facilitate scalability, and enable new feature development
• Introduced a CI pipeline with automated linting, testing, and non-production environments, leading to lower incident rate and code churn
• Implemented observability using OpenTelemetry, Sentry, and Google Cloud tooling, utilizing profiling, tracing, and logs to debug and fix production incidents
• Added SSO and RBAC for users to meet requirements for SOC 2 and HIPAA compliance

highlights

• Created vendor-facing REST APIs used to query and load data in near real time instead of daily/weekly batches using Java Spring and Oracle SQL
• Re-architected Java and Informatica applications, improving performance by up to 50%
• Created a library and utility in Go to parse and validate Informatica workflows
• Developed a program to validate production deployments and sync environments using Python, reducing manual work and improving accuracy, saving over 30 hours per release
• Formed a collaborative team knowledge base for onboarding and team documentation using Confluence
• Piloted the transition of the business to Agile from Waterfall

highlights

• Developed features and debugged software written in C++, C#, and Javascript, and supported client IT teams with on-site devices
• Analyzed millions of call detail records (CDR) using Bash/Python scripts and Tableau to help clients pinpoint issues in their data

highlights

• Designed and developed client websites (WordPress, CakePHP, Symfony, jQuery) while managing the system and network infrastructure that ran client applications, websites, and email