An overview of this role
As a Staff Backend Engineer (Python) on the Duo Chat team in AI Engineering, you'll lead the backend architecture that powers GitLab Duo Chat across the GitLab DevSecOps platform. You'll solve hard problems in building reliable, secure, and scalable AI-powered chat workflows so customers can plan, write, review, and secure code faster, with confidence. This is a hands-on technical leadership role where you'll set direction for how we integrate and evolve large language model providers (including Google Vertex AI) across Ruby on Rails and Python services, raise the bar on observability and testing, and guide the team through ambiguous, high-impact technical decisions. Over your first year, you'll be expected to drive key architectural choices, reduce technical debt that slows iteration, and help the team ship durable improvements to response quality, reliability, and maintainability.
Some examples of our projects:
- Integrate new generative AI models and providers into GitLab Duo Chat to expand capabilities and improve response quality
- Improve debugging, observability, and test coverage for AI-powered chat workflows to increase reliability at scale
What you'll do
- Define the technical architecture and technical roadmap for the Duo Chat group, aligning backend execution with product direction and engineering priorities
- Solve the highest-scope and most ambiguous backend problems, delivering secure, well-tested, performant solutions with minimal guidance
- Integrate and extend generative AI capabilities in GitLab Duo Chat, including large language models (LLMs) and providers such as Google Vertex AI
- Develop, ship, and maintain backend features across Python and Ruby on Rails services that power Duo Chat experiences across the GitLab platform
- Design, implement, and review GraphQL application programming interface (API) contracts and supporting backend logic to ensure reliability, scalability, and clear frontend integrations
- Improve observability, debugging workflows, and incident readiness by strengthening logging, tracing, and production troubleshooting practices
- Drive code quality and long-term maintainability by setting internal standards, leading code reviews, and identifying and reducing technical debt
- Mentor engineers across the team and participate in Tier 2 on-call rotations, contributing to root cause analysis and follow-up improvements to resiliency and testing (including RSpec)
What you'll bring
- Production experience building and operating backend services in Python, including background jobs, APIs, and data models
- Ability to define and evolve technical architecture by weighing trade-offs, selecting patterns and tools, and setting a clear technical direction for others to follow
- Experience setting and driving a technical roadmap in partnership with product and engineering stakeholders
- Proficiency designing and maintaining REST and/or GraphQL APIs with attention to scalability, maintainability, and backward compatibility
- Hands-on experience integrating large language models into applications, including prompt design and building features powered by generative AI
- Strong SQL skills and experience working with relational databases such as PostgreSQL, including efficient queries and data modeling
- Experience mentoring engineers through code review, architectural guidance, and shared standards, and communicating complex technical decisions in a clear, async-first way
- Comfort contributing in a mature codebase across Python and Ruby on Rails, with openness to learning and applying transferable skills from related technologies or domains
About the team
The Duo Chat team sits within GitLab's AI Engineering organization and is responsible for building and evolving GitLab Duo Chat, the AI-powered chat experience embedded across the GitLab DevSecOps platform. You'll work with a small, cross-functional group of backend, frontend, and AI specialists who collaborate asynchronously across time zones, using GitLab issues, merge requests, and documentation as the primary way of working. The team focuses on integrating and scaling generative AI capabilities (including providers like Google Vertex AI), improving reliability and performance, and strengthening debugging, observability, and testing workflows so customers can safely use AI to plan, write, review, and secure their code across GitLab.