Generative AI Engineer

Position Description

We are seeking a Generative AI Engineer to own the hands-on technical delivery of production GenAI systems - from architecture and implementation through deployment, operations, and continuous improvement. This role covers the full stack of modern GenAI engineering: LLM application design, agentic and RAG workflows, structured output patterns, evaluation pipelines, and operational safeguards, integrated with enterprise data sources and cloud-native services.

General Duties and Responsibilities

• AI Architecture & Delivery
  • Design and build production-grade generative AI systems - agentic workflows, multi-step RAG pipelines, and LLM-powered applications integrated with enterprise data and services
  • Define and implement reusable engineering patterns for prompt management, workflow versioning, structured outputs, tool orchestration, and rollback across production AI services
  • Apply judgment around model selection and routing, token and latency optimization, cost management, and the appropriate boundaries between AI-driven and deterministic application logic
  • Continuously evaluate emerging AI models, tools, and architectural approaches, incorporating improvements into existing systems incrementally
  • Integrate AI systems with enterprise data sources, internal APIs, and platforms to enable reliable, production-ready workflows
• Reliability, Performance & Operations
  • Own operational outcomes for production AI systems - reliability, latency, throughput, cost efficiency, and scalability targets
  • Implement and maintain monitoring, observability, tracing, and alerting frameworks to ensure operational visibility and rapid issue resolution
  • Lead production incident response and root cause analysis, driving systemic improvements that reduce recurrence
• Governance, Security & Responsible AI
  • Build and maintain automated evaluation pipelines for LLM outputs - prompt regression testing, retrieval quality validation, and failure mode tracking
  • Implement human-in-the-loop controls, content guardrails, schema validation, and structured output enforcement to ensure trusted and auditable AI outputs
  • Secure AI systems against prompt injection, data leakage, and unauthorized access, aligning with enterprise compliance and security standards
• Technical Authority & Collaboration
  • Own the team's GenAI technical direction - defining and enforcing engineering standards, patterns, and best practices across all GenAI workstreams
  • Make and defend architectural decisions with clarity, providing the technical rationale needed for the Manager and stakeholders to align and move forward confidently
  • Work closely with the Manager, GenAI Engineering to receive, refine, and execute on scoped GenAI work - contributing technical judgment to prioritization and tradeoff decisions
  • Provide hands-on code review and technical guidance to engineers contributing to GenAI workstreams, raising overall quality through direct feedback and demonstration
  • Champion an iterative delivery culture - shipping incrementally, incorporating feedback, and improving continuously in a regular production release cadence

Education and/or Experience

• Required:
  • Demonstrated experience shipping production-grade LLM or generative AI systems - prompt and workflow design tradeoffs, model selection and routing decisions, tool use and agent orchestration boundaries, and the distinction between AI guardrails and deterministic application logic
  • Experience building automated evaluation pipelines for LLM outputs, including gold set construction, model-based evaluation approaches, prompt regression testing, retrieval quality validation, and failure mode analysis across the full LLM application stack
  • Strong track record designing, building, and operating complex distributed systems in enterprise production environments, with clear ownership of reliability, performance, and operational outcomes
  • Proven ability to define and enforce GenAI engineering standards, patterns, and best practices across a cross-functional team
  • Experience with CI/CD pipeline design and operation for AI services - including deployment strategies, versioning, and release management in production environments
  • Experience integrating AI systems with enterprise data sources, internal APIs, and security controls in compliance-sensitive environments
  • Demonstrated track record of shipping production AI systems iteratively - with regular release cadence, feedback incorporation, and continuous improvement
• PREFERRED:
  • Experience designing and operating agentic AI systems and multi-step RAG architectures in production - retrieval quality optimization, chunking strategies, grounding, and ranking tradeoffs
  • Hands-on experience with Azure OpenAI, AI Foundry, App Service, Functions, Service Bus, Blob Storage, Key Vault, and Application Insights; familiarity with Bicep for IaC
  • Familiarity with PySpark notebooks for data pipeline development
  • Familiarity with deploying and managing containerized AI workloads using Docker or similar technologies
  • Familiarity with responsible AI principles, AI governance frameworks, and regulatory considerations relevant to enterprise AI systems
  • Familiarity with Bronze/Silver/Gold medallion architecture and staged data quality patterns for enterprise data pipelines
  • Domain experience in product data, PIM, ERP, master data management, data governance, ecommerce, or analytics platforms

Physical Job Requirements

• Prolonged periods in a stationary position at a desk and working on a computer
• Must be able to communicate effectively via video conferencing, phone, and written correspondence
• Occasional travel may be required depending on project or business needs

Work Environment

• The work environment is typically in a remote office setting during normal or extended business hours
• Candidates for the position should be able to perform essential job duties in described work environment with or without accommodation
• Reasonable accommodations may be made to enable individuals with disabilities to perform the essential functions