Technical Architect - AI

Location : Santa Clara, CA Job Type : DTH/Contract Job Description: 10–12 years in AI/ML, Cloud, platform engineering, or enterprise architecture roles 3+ years of hands-on experience with GCP Vertex AI, including real work with components such as Agentspace, Agent Builder, Vector Search (Matching Engine) amd/or Search & Conversation 1 year hands-on experience with Agent Development, Agentspace, Agent Builder, Search & Conversation; able to describe at least one actual solution built on Vertex AI Prior experience designing enterprise-grade AI, GenAI or Agentic systems including aspect of Agent Ops and Ob Behalf of Workflows; Exposure to multi-cloud AI environments (Azure OpenAI, Copilot Studio, OpenAI API) 2+ years of experience building Generative AI applications, such as AI assistants, retrieval-based systems, or LLM-powered workflows. The candidate should clearly explain what they built and what their role was 5+ years of strong Python development experience, specifically building backend services, APIs, Microservices, or automation components used in production environments Practical integration experience with at least one enterprise platform (SAP, Salesforce, or ServiceNow), with the ability to describe a real integration scenario they worked on 3+ years of cloud deployment experience, preferably using GCP services like Cloud Run, Cloud Functions, or Kubernetes for deploying and maintaining cloud-native applications 1–2 years of experience operationalizing AI systems, including managing prompts or models, handling errors or failures, monitoring performance, or improving system reliability. Exposure to LLMOps or similar processes Basic working knowledge of enterprise security and data protection, including responsible handling of sensitive data, access control, and safe use of AI systems in an enterprise environment Bachelor’s or Master’s degree in Computer Science, Data Science, Engineering, or a related technical discipline Strong communication skills, with the ability to explain past projects clearly, walk through their contributions, and provide understandable examples of their AI and cloud experience. The GCP Vertex AI + Agentspace architecture Standards for agent development patterns, grounding, and memory Integration of agents with SAP/Salesforce/ServiceNow A2A (Agent-to-Agent) coordination and orchestration design Context Engineering patterns for reliable grounding Approaches for testing, observability, safety, and control in GenAI systems Enterprise governance, LLMOps, AgentOps, and lifecycle management Primary Responsibilities: Architect Scalable Vertex AI & Agentspace Solutions – Design and deliver AI architectures built on GCP Vertex AI and Agentspace, covering agent workflows, retrieval pipelines, vector search, grounding logic, tool integrations, and multi-agent (A2A) coordination; Ensure the platform is secure, resilient, and built for scale Platform Strategy & Technical Direction – Provide guidance on architecture patterns, technology choices, and platform evolution; Help teams understand trade-offs, make decisions that align with long-term business outcomes RAG Systems & Context Engineering – Lead the design of retrieval pipelines and context strategies that produce reliable, high-quality responses; Define how data is chunked, embedded, searched, and assembled into grounded context windows for agents Agentic AI Frameworks & A2A Patterns – Define patterns for building and coordinating agents across Agentspace, LangGraph, DSPy, or similar frameworks; Establish approaches for delegation, task planning, error recovery, and safe inter-agent communication Tooling Integration & MCP-Style Interfaces – Architect how agents call tools and external systems; Define tool schemas, safety constraints, validation rules, and execution boundaries across SAP, Salesforce, ServiceNow, and enterprise APIs LLMOps & AgentOps – Set up operational foundations for prompts, models, and agents — including CI/CD pipelines, monitoring dashboards, version control, error tracking, and cost governance; Implement guardrails that reduce hallucinations and prevent unsafe or unintended behavior Design Authority & Governance – Lead architecture reviews, define reference architectures, and establish reusable patterns. Ensure every GenAI initiative adheres to security, data governance, and platform standards Cross-Functional Collaboration – Work closely with engineering, data, product, and business teams to convert use cases into practical, production-ready architecture. Break down complexity so teams can execute confidently Documentation & Standards – Create and maintain playbooks, best practices, design guides, and reference implementations that help distributed teams build consistently Monitoring, Testing & Observability – Establish testing frameworks for retrieval quality, agent behavior, grounding accuracy, and safety signals; Guide development of AgentOps dashboards that track performance, tool failures, latency, drift, and system health. Secondary Responsibilities: Platform Research & Innovation – Stay current with advancements across Vertex AI, Agentspace, Model Garden, and broader agentic patterns. Bring forward ideas worth evaluating and scaling Proof of Concepts – Lead/sponsor PoCs to validate feasibility, performance business value before full-scale adoption Ecosystem Awareness – Maintain familiarity with Azure OpenAI, Copilot Studio, AI Studio, Cognitive Search, and other cloud AI platforms to support multi-cloud strategy Business Alignment – Engage with product and business leaders to identify impactful use cases, help shape roadmaps, and clarify expected outcomes Mentorship & Skill Building – Support engineers through coaching on RAG tuning, prompt refinement, agent patterns, testing techniques, and responsible AI practices