Explicitly references "vibe coding" and requires mentoring engineers to adopt vibe coding practices for agentic development.
About the Role
Lead design and productionization of autonomous, agentic AI systems to automate and optimize internal engineering workflows. Drive architecture, integration, and delivery of Gemini-powered agents and cloud-native services while mentoring engineers and liaising between product and engineering stakeholders.
Job Description
Role
Lead the design, development, and productionization of autonomous, task-oriented AI agents to optimize internal engineering workflows, automate complex problem solving, and improve platform efficiency. Act as an architectural and delivery lead from requirements gathering through production release, and mentor junior engineers in agent development practices.
Key Responsibilities
- Architect and orchestrate agentic workflows capable of planning, executing, and verifying engineering tasks (e.g., code generation, unit test creation, automated bug triaging).
- Develop and maintain Model Context Protocol (MCP) servers to connect agents to canonical documentation and tools to ensure grounded responses.
- Build context-aware assistants using Agent Development Kit (ADK) and integrate them with IDEs and development tools via the Google Antigravity platform.
- Lead productionization of AI POCs into production-ready services in collaboration with search and productivity teams.
- Propose architectural improvements to enhance reasoning capabilities, reduce latency, and optimize token usage.
- Provide technical mentorship and guide adoption of agentic tools and “vibe coding” practices.
Requirements
- Minimum 8 years professional software engineering experience with a proven record in full‑stack or backend development.
- Strong practical experience in Java and Python.
- Hands-on experience with large language models (LLMs), specifically the Gemini family, and applying them in agentic workflows, reasoning tasks, and tool-calling.
- Familiarity with Agent Development Kit (ADK), Model Context Protocol (MCP), and agent design within Vertex AI.
- Strong understanding of cloud-native development on Google Cloud Platform (GCP).
- Demonstrated ability to lead complex projects and act as a strategic liaison between engineering and product stakeholders.
Nice to Have
- Experience with Google-proprietary technologies such as Borg, Blaze, Piper, and Critique.
- Experience with GCP services such as Spanner, BigQuery, and Pub/Sub.
Tech / Tools Mentioned
Java, Python, Gemini (LLM family), Agent Development Kit (ADK), Model Context Protocol (MCP), Vertex AI, Google Cloud Platform (GCP), Borg, Blaze, Piper, Critique, Spanner, BigQuery, Pub/Sub, Google Antigravity platform, Moma, IDE integrations.