Senior AI Engineer, Shine

• 6+ years of software engineering experience with strong production engineering fundamentals (shipping and operating real systems)
• Thorough, current, hands-on expertise in AI-native engineering: agents, tool use, multi-step workflows, RAG, context- and harness-engineering, including common failure modes and dependability patterns
• Practical experience with eval and observability frameworks for non-deterministic, LLM-powered systems
• Deep familiarity with everyday product engineering workflows (code review, CI/CD, incident handling, delivery pressure)
• Strong collaborative mindset (teamplay as a working style): comfortable in other teams' codebases, giving/receiving direct feedback
• Strong delivery orientation: iterative shipping, preferring working v1 over perfect plans, with accountability for long-term outcomes
• Excellent technical communication across all engineering levels

Desirable

• Shine is moving all of Product & Tech toward AI-native engineering. The Enablement team builds the platform, workflows, and standards that make this vision real—placing tools directly into the hands of engineers
• Working under the guidance of program technical leadership, the team turns AI-native SDLC concepts into working software, reference implementations, and widely adopted practices
• You are where the AI-native platform gets built and lands in the hands of engineers. You will work closely embedded with product engineering teams—close enough to how they build software that your tooling, patterns, and support never feel alien or imposed from the outside
• This is a senior individual contributor role with real delivery ownership. While technical direction is set at the program level, bringing it to life, making it reliable, and ensuring high adoption rests substantially on you
• Platform components, end-to-end: Build and maintain concrete parts of the AI-native SDLC platform (agent workflows, MCP servers, skills, configuration presets, eval harnesses, developer tooling). Deliver production-grade components with clear ownership, quality gates, and lifecycles
• Hands-on Platform Engineering (~80% of time): Write production code for agent infrastructure, tooling integrations, MCP servers, and eval pipelines
• Team Adoption & Pairing: Embed with product engineering teams in their codebases to introduce AI-native workflows, debug friction firsthand, run enablement sessions/workshops, and translate blockers into platform improvements
• Working Patterns & Reference Implementations: Produce credible, working examples in real repositories so teams adopt patterns by copying what demonstrably works
• Quality, Evals & Observability: Build eval harnesses, observability, and verification for non-deterministic systems to make agent behavior trustworthy before reaching teams
• Engineering Standards & Frontier Research: Contribute to shared coding conventions, trust/verification policies, and presets while evaluating new models, tools, and protocols (MCP, A2A, agent frameworks)
• How We Measure Success:
• Delivery: Platform components shipped to production quality with evals and observability; trending down rework rates and incidents
• Adoption: Active weekly use of platform workflows by your supported teams; reduced time-to-productivity for adopting new AI-native workflows
• Trust: High DX survey satisfaction, trust in agent-generated work, and engineering teams actively pulling you in rather than platform push

• We usually respond within a week