Sr. Staff Enterprise AI Architect in Redwood City
Role details
Job location
Tech stack
Job description
As a senior architectural leader within the Enterprise AI Hub, you will design and implement scalable AI systems aligned to Reltio's enterprise AI strategy. You will partner with the Director, Enterprise AI to evolve the foundational architecture that enables secure, governed agentic workflows across the organization-helping to engineer the "Reltio Brain" (our Context Intelligence Operating System).
In this role, you will evaluate cloud AI platforms, orchestrate vector databases, and help build the Model Context Protocol (MCP) gateways that securely connect AI to our corporate ecosystem. You will spend a meaningful portion of your time working directly with specialized AI Engineers to build the foundational primitives ("building blocks") that empower embedded AI Business Partners to innovate safely and rapidly.
Job Duties and Responsibilities:
Agentic Architecture & Orchestration:
- Lead the application of advanced principles to design and develop the overarching architecture for multi-agent systems, moving beyond simple conversational RAG to deterministic, goal-driven planning loops.
- Define how specialized agents communicate, delegate tasks, and maintain persistent state across complex business workflows using frameworks like LangGraph or CrewAI.
Model Context Protocol (MCP) Governance:
- Architect centralized MCP gateways.
- Standardize how all internal agents securely discover, authenticate, and connect to enterprise systems (Salesforce, Reltio-on-Reltio, HRIS, Jira).
- Eliminate bespoke API integrations in favor of a universal, auditable tool-calling ecosystem.
Semantic Layer & Vector Infrastructure:
- Enable the enterprise data grounding strategy with infrastructure design.
- Establish advanced patterns for document ingestion, semantic chunking, and embedding model selection across the L1 (Personal), L2 (Team), and L3 (Enterprise) context hierarchy.
- Select and optimize enterprise-grade vector databases (e.g., Pinecone, Milvus) for latency and recall accuracy.
Deterministic Security & Access Control:
- Implement rigid, deterministic software layers to govern probabilistic LLM outputs.
- Design robust Role-Based Access Control (RBAC) and Attribute-Based Access Control (ABAC) mechanisms within the MCP layer to ensure strict data isolation and prevent sensitive context leakage or prompt injection attacks.
Continuous Evaluation & Cost Optimization:
- Partner with engineering leadership in the implementation and maturation of the Agentic Software Development Life Cycle (SDLC)
- Implement continuous Agentic Evaluation (Evals) harnesses and LLM-as-a-Judge frameworks to monitor agent drift, hallucination rates, and silent failures in production.
- Design intelligent model routing strategies (balancing frontier LLMs with Small Models) to optimize inference costs and throughput.
Technical Leadership & Enablement:
- Provide decisive architectural leadership, guiding complex technical decisions and ensuring alignment to enterprise standards.
- Mentor a highly specialized pod of AI Engineers, enforcing rigorous systems-engineering standards.
- Create formal networks and coordinate among groups within the organization, conveying AI capabilities to diverse stakeholders and assisting in translating them into measurable business ROI.
Requirements
- Typically requires a minimum of 12 years of related experience with a Bachelor's degree; or 8 years and a Master's degree; or a PhD with 5 years experience; or equivalent experience. This must include progressive software architecture and distributed systems engineering, with a minimum of 2+ years exclusively focused on production-grade generative AI, LLM orchestration, and autonomous agent deployment.
- Deep, practical mastery of the Model Context Protocol (MCP) and its implementation for secure, universal enterprise tool calling.
- Advanced expertise in multi-agent orchestration frameworks (e.g., LangGraph, CrewAI, Semantic Kernel) and transitioning from basic scripting to fault-tolerant system design.
- Authoritative knowledge of vector databases (Pinecone, Weaviate, Milvus), hybrid search architectures (dense/sparse retrieval), and advanced RAG optimization techniques at scale.
- Strong security engineering mindset with the ability to build deterministic guardrails, data masking, and management (Okta/OAuth/RBAC/ABAC) over non-deterministic systems.
- Command of modern AI observability and MLOps, including CI/CD for non-deterministic systems and tracing tools like LangSmith.
- Extreme adaptability and a "builder" ethos: high tolerance for ambiguity and the agility to rapidly pivot architectures when the underlying foundation models or industry protocols evolve.
Skills That Are Nice to Have:
- Experience in a 500-2,000 employee cloud- SaaS environment.
- Experience working closely with core MDM, data engineering, or analytics teams.
- Working knowledge of key business systems such as Salesforce, NetSuite, or Jira from a data-model and API-integration perspective.
- Familiarity with local LLM deployment, SLM routing, or sandbox environment virtualization.
- Prior involvement in setting up automated LLM evaluation pipelines (Evals).
Benefits & conditions
At Reltio, we carefully consider a wide range of compensation factors to determine your personal top of market. We rely on market indicators to determine compensation and your specific job family, background, skills, and experience to get it right. These considerations can cause your compensation to vary and will also be dependent on your location. Overall Market Range$132,000-$244,000 USD
Reltio is proud to be an equal opportunity workplace. We are committed to equal employment opportunity regardless of , , ancestry, , , , , , citizenship, marital status, , or Veteran status. We also consider qualified applicants regardless of criminal histories, consistent with legal requirements. Reltio is committed to working with and providing reasonable accommodation to applicants with physical and mental disabilities.
