Distinguished Engineer, Agentic SDLC & Non‑Linear Productivity

The role

This is the most senior hands-on technical IC position, focused on applying autonomous AI agents to GitLab's own development lifecycle first, then productizing what proves effective. The role connects architecture, product, infrastructure, and data/ML groups and reports to a director-level engineering leader.

What you would do

• Set a company-wide technical direction for self-directed AI agent work in software delivery, kept consistent with product plans and engineering role standards.
• Look for lifecycle areas, from planning and coding through review, security, compliance, and operations, where agents could take over most of the work and deliver order-of-magnitude gains.
• Convert ambiguous problem areas into staged, revisable plans with product, AI/ML, and architecture partners.
• Build and test prototypes for tasks such as generating merge requests, triaging tests, fixing vulnerabilities, release readiness checks, and incident handling.
• Draft reference designs covering orchestration, protective limits, monitoring, and points where a human must approve agent actions.
• Establish offline benchmarks and live experiments to gauge accuracy, speed, safety, cost, and effect on productivity.
• Shepherd a handful of internal pilot efforts from initial idea through adoption to measurable results.
• Help engineering teams fold agent workflows into daily work, and track delivery metrics such as cycle time and recovery time.
• Package repeatable patterns, libraries, and guides so other teams can adopt them with little effort.
• Partner with product and engineering leadership to move proven internal patterns into safe, reliable customer features that respect tenant isolation and data rules.

What they asked for

• Ten or more years of software engineering work, including at least four years in a staff-level or comparable senior technical leadership role.
• Hands-on command of large language model systems and multi-step agent frameworks running at real production scale.
• Track record of running rigorous experiments, building evaluation suites, and turning findings into architecture decisions.
• Background designing distributed platforms that serve many isolated tenants, including failover, telemetry, and recovery behavior.
• Experience building approval checkpoints, safety limits, and responsible-AI safeguards into live systems.
• Ability to align engineering, product, infrastructure, and data/ML groups on ambiguous, high-stakes technical problems.
• History of coaching senior engineers and steering technical direction across teams without formal authority.
• Comfortable working in a fully distributed company and communicating clearly in writing and asynchronously.

Nice to have

• Practical grounding in DevSecOps tooling, CI/CD building blocks, and software delivery lifecycles.
• Public speaking or open-source contributions on AI-assisted development or autonomous agents.