Matthew-Luke Marion

Senior Software Engineer · Infrastructure & AI Productivity

I'm a senior software engineer on the infrastructure team at SurveyMonkey, and the person the organization comes to for AI productivity work.

Most of what I build now is internal. Recently, a code intelligence service on Amazon Bedrock Knowledge Bases that indexes every repository we have and answers questions about them through an MCP server, and the Slack assistants that sit on top of it.

Before that, five years of product engineering, much of it on security and incident response. I've worked incidents as the lead, as the subject-matter expert, and as the engineer who found and fixed the problem.

I came into engineering the long way. Ten years ago I was answering support tickets for FluidReview. Then I tested the product, then I built it, and now I build the tools the engineers building it depend on. No CS degree, self-taught throughout.

I also run End of Line Software, a small Canadian consultancy.

Languages
Python · TypeScript · JavaScript · SQL
Applied AI
Amazon Bedrock · Retrieval-augmented generation · Vector search · Model Context Protocol · Agentic tool use · Prompt and output-schema design
Platform
AWS · Kubernetes · Docker · Terraform · GitHub Actions · GitOps with ArgoCD
Data & Services
PostgreSQL · Redis · RabbitMQ · REST APIs · Django · React
Practice
Incident command · Threat modelling · Disaster recovery planning · Automated testing · Observability · Mentorship

Senior Software Engineer, Infrastructure — SurveyMonkey

September 2025 – Present

Internal tooling and the AI systems the engineering organization runs on.

  • Built an organization-wide code intelligence service on Amazon Bedrock Knowledge Bases, covering 1,200+ repositories across 50+ GitHub organizations.
  • Seeded the index with a baseline pass over the whole estate, then built a GitHub Actions pipeline that re-ingests on merge to main, keeping answers current within about fifteen minutes.
  • Attached ownership metadata during ingestion so results name the owning team and its contact channel alongside the code.
  • Built the query surface as a FastMCP server: cross-repository search, example-usage lookup, impact analysis, downstream dependency review, and breaking-change detection for callers of a changed interface.
  • Automated MCP provisioning through an internal CLI that already handles AWS authentication, so it self-configures in engineers' local AI tooling. Every engineer in the organization now uses it, directly or through the infrastructure assistant.
  • Built a Slack assistant answering engineering questions from Jira, Confluence, OpsLevel, and source, and instrumented it to keep model spend visible and attributable.
  • Added an agentic loop and gave it the code intelligence MCP server as a tool, so one question can cross documentation, ticket history, service ownership, and live code analysis.

Senior Software Engineer — SurveyMonkey

March 2024 – September 2025

Product engineering with a heavy security and reliability focus.

  • Worked security incidents alongside the security and compliance teams. Sometimes incident lead, sometimes subject-matter expert, often the engineer who found and fixed the problem.
  • Built disaster recovery and business continuity plans with compliance.
  • Threat-modelled features before implementation, and designed security tooling and workflows that made audits routine instead of disruptive.
  • Designed and built accessible React interfaces, replacing an outdated UI.
  • Ran office hours and training sessions, and co-authored onboarding modules with the enablement team.

Software Engineer II — SurveyMonkey

August 2022 – March 2024

  • Led Apply's move to GitHub Actions for CI/CD, cutting deployment time and stabilizing builds.
  • Rewrote plan-management tooling so sales could differentiate plans and customize them without engineering help.
  • Worked with customer success and implementation on retention issues.

Software Engineer I — SurveyMonkey

August 2021 – August 2022

  • Migrated the product from legacy infrastructure to AWS.
  • Fixed performance and reliability problems that had outlasted several earlier attempts.
  • Joined the on-call rotation and standardized the release process.

Software Test Engineer II — SurveyMonkey

September 2019 – August 2021

  • Wrote test automation for new features and raised coverage on existing ones.
  • Containerized the product for development, testing, and production use.