What are the responsibilities and job description for the Google Cloud Platform AI Architect Diamond Layer Tech Lead_New Jersey,NJ position at Keylent?
Google Cloud Platform AI Architect (Diamond Layer Tech Lead)*
Location - NJ/ATL
Role Overview
We are seeking a Diamond Layer Tech Lead to lead the architecture and implementation of enterprise AI orchestration, domain agents, semantic retrieval, knowledge graphs, and supporting data foundations.
This role will own the technical architecture connecting the Master Coordinator, domain agents, tools, RAG systems, semantic platforms, and enterprise data sources, with a strong emphasis on grounded, accurate, traceable, and production-ready AI systems.
The ideal candidate combines deep hands-on expertise in agentic AI, RAG, knowledge graphs, Python, and data engineering with the technical leadership skills required to establish reusable enterprise AI patterns.
Key Responsibilities
• Own the architecture of the Master Coordinator, domain agents, semantic retrieval, knowledge graphs, and supporting data foundations.
• Build production-grade agent services using advanced Python.
• Design multi-step agent workflows, orchestration state, tool invocation, fallback handling, and reusable agent patterns.
• Establish standardized agent and tool contracts, agent registration, and model/platform adapters.
• Design deterministic and LLM-assisted routing across multiple enterprise domains.
• Build data foundations using advanced SQL, BigQuery, and GCS.
• Perform source discovery, data-gap analysis, business-key validation, reconciliation, data-quality assessment, and source-of-truth determination.
• Establish data freshness, lineage, quality, and governance practices.
• Design and optimize RAG pipelines, including retrieval, reranking, grounding, prompt/context assembly, citation support, and hallucination reduction.
• Enable reliable cross-domain information retrieval and synthesis.
• Design and maintain enterprise knowledge graphs and semantic models using RDF, SPARQL, SHACL, Stardog, and ontology modeling.
• Develop virtual graphs and relational-to-semantic mappings and establish semantic versioning and graph-promotion processes.
• Deploy and operate agent runtimes on Google Cloud Platform/GKE and integrate them securely with BigQuery, GCS, and semantic platforms.
• Establish CI/CD processes for agents and knowledge graphs, including version-controlled prompts, tools, mappings, ontologies, queries, and deployment definitions.
• Develop automated evaluation gates for AI releases.
• Build comprehensive agent evaluation frameworks measuring accuracy, relevance, groundedness, completeness, hallucination, latency, and cost.
• Develop benchmark scenarios and gold-answer datasets for regression testing.
• Implement end-to-end observability across coordinator, agent, tool, semantic, and data layers using Langfuse or equivalent LLM observability platforms and Grafana.
• Implement identity-aware retrieval, least-privilege data access, cross-domain guardrails, prompt/data protection, source attribution, and end-to-end auditability.
• Lead architecture and design reviews, remain hands-on with implementation, mentor engineers, and establish reusable agent, semantic, and data patterns.
Required Qualifications
• Advanced hands-on Python development experience.
• Strong experience building LLM agents, agentic workflows, and orchestration systems.
• Deep expertise in RAG, including retrieval, reranking, grounding, context construction, and hallucination mitigation.
• Strong experience with RDF, SPARQL, SHACL, ontology modeling, and knowledge graphs.
• Experience with Stardog or comparable enterprise semantic platforms.
• Advanced SQL and BigQuery expertise.
• Experience with data lineage, reconciliation, quality, and source-of-truth assessment.
• Experience deploying AI/agent workloads using Google Cloud Platform and GKE.
• Experience implementing LLM/agent evaluation frameworks and automated quality gates.
• Experience with Langfuse or equivalent LLM observability tools and Grafana.
• Strong understanding of enterprise AI security, governance, identity-aware retrieval, and auditability.
• Demonstrated ability to lead technical architecture, establish reusable engineering patterns, conduct technical reviews, and mentor engineers.