Senior AI Engineer

Senior AI Engineer

Location: Santa Clara - Hybrid

• Design and deploy multi-agent systems that utilize reflection, self-correction, and planning loops to solve open-ended business problems.

• Build complex, cyclic AI workflows using LangGraph to handle long-running tasks, state persistence, and error recovery.

• Implement and fine-tune Gemini (2.5 Pro/Flash) models within the Vertex AI ecosystem, optimizing for latency, token efficiency, and reasoning quality.

• Develop secure, robust interfaces for agents to interact with external APIs, databases (SQL & Vector), and legacy enterprise systems.

• Architect "Agentic RAG" systems that don't just retrieve data, but intelligently route queries, verify source relevance, and synthesize multi-step answers.

• Set up advanced monitoring for agentic behavior, including tracing "chain-of-thought" reasoning, cost tracking, and automated evaluation of agent performance.

Skills:

• 7+ years of experience with Python, with deep expertise in asynchronous programming and building scalable APIs (FastAPI/Flask).

• Proven track record of shipping production-grade applications using LangGraph

• Expert-level experience with Vertex AI, including Model Garden, Vertex AI Search & Conversation, and deploying models via Vertex Endpoints.

• LLM Expertise, Deep understanding of Gemini capabilities (function calling, multimodal input, long-context window management).

• Hands-on experience with Vertex AI Vector Search (formerly Matching Engine) or similar databases like Pinecone/Weaviate.

• Strong grasp of CI/CD for AI, containerization (Docker/GKE), and "eval-driven development" to measure agent accuracy.

• Experience with Model Context Protocol (MCP) for standardized tool-calling.

• Familiarity with alternative agent frameworks like CrewAI or Semantic Kernel.

• Contributions to open-source AI libraries or a strong portfolio of agentic research projects.