Senior AI Architect & Engineer
Location: Remtoe
Job Summary
We are seeking a Senior AI Architect & Engineer to serve as a technical partner to the VP of Enterprise Architecture & Engineering. This is a senior-level role for someone who operates effectively across architecture and engineering — contributing to integration pattern design, building and delivering AI pilots, and working alongside platform engineering teams to scale validated capabilities into production.
This role supports five concurrent AI programs: an enterprise AI platform migration, an AI-powered attendee intelligence pilot, a synthetic workforce platform, an AI-enabled software delivery methodology, and a proposal automation capability. The posture shifts by project phase — contributing to architecture design when defining integration patterns, building prototypes when proving those patterns against real systems, and working embedded with delivery teams when scaling to production.
The ideal candidate brings strong, current experience in large language model application development, AI agent implementation, enterprise RAG pipelines, and API-based integration patterns. They understand how to move AI out of proof-of-concept and into governed, observable enterprise production — and have done it before. They work effectively within enterprise governance frameworks and produce architecture artifacts that engineering teams can build from.
Key Responsibilities
AI Architecture & Pattern Implementation
• Implement and extend Freeman's MCP (Model Context Protocol) integration pattern — connecting enterprise AI platforms to business data sources including event APIs, CRM, ERP, data warehouse, and document repositories.
• Contribute to AI integration framework design that governs how large language models, RAG pipelines, agent systems, and API layers connect to Freeman's core platforms.
• Design and implement AI agent patterns including multi-agent orchestration, tool-use, memory architecture, and human-in-the-loop oversight for assigned programs.
• Support the AI-Enabled Agile Delivery framework — contributing to workflow design, role-based AI SOP libraries, and tooling that compresses software delivery timelines.
• Produce architecture decision records (ADRs), integration specifications, and NFR documentation that meet ARB governance standards and serve as references for engineering teams.
• Contribute to data governance patterns including prompt versioning, evaluation frameworks, and production readiness checklists for AI workloads.
Prototype & Pilot Delivery
• Build working AI prototypes that prove integration patterns against real enterprise systems — instrumented, documented, and built to be transferable to delivery teams.
• Own pilot delivery from architecture validation through production-ready handoff — enforcing NFR compliance and ensuring operational readiness before scale.
• Implement MCP server integrations connecting enterprise AI platforms to business data sources using extensible patterns that delivery teams can build on.
• Engineer RAG pipelines with documented retrieval quality metrics, chunking strategies, embedding model selection rationale, and latency targets validated against real content.
• Build AI agent implementations with observable behavior, job-scoped tool access, and governance documentation sufficient for ARB review.
• Develop AI-Enabled Agile Delivery tooling — prompt libraries, project configurations, and workflow SOPs — that delivery teams can adopt independently.
Production Scale Support
• Work embedded alongside platform engineering teams to transition validated prototypes into production-grade capabilities — providing technical guidance and pattern enforcement during scale-up.
• Build repeatable deployment patterns, operational runbooks, and extension guidelines that allow permanent engineering teams to maintain and extend AI capabilities.
• Instrument production AI workloads with cost tracking, performance monitoring, model drift detection, and explainability documentation.
• Partner with the Cloud Center of Excellence on FinOps tagging and cost attribution for AI workloads running on cloud platforms.
• Identify rationalization opportunities across the AI toolchain and flag platform sprawl before it creates technical debt.
Governance & Standards Contribution
• Contribute to Architecture Review Board (ARB) processes as a technical participant for AI platform, agent, and integration decisions.
• Support the Data & AI Working Group's architecture standards evolution by producing integration specifications and pattern documentation.
• Apply Zero Trust security principles to AI architecture designs — securing agent communications, MCP server interactions, and data access patterns by design.
• Maintain architecture artifacts as living references — ADRs, NFR catalogs, and integration specs must reflect production reality, not point-in-time designs.
Use Case Development
• Translate business requirements into concrete AI capability designs mapped to core value streams: Quote-to-Cash, Idea-to-Market, and Plan-to-Inventory.
• Build proof-of-concept demonstrations for executive and client audiences that validate technical differentiation with supporting documentation.
• Collaborate with innovation, product, and business stakeholders to identify and scope AI use cases by feasibility, value, and strategic alignment.
Key Skills & Qualifications
Technical Skills
1. MCP & Enterprise AI Integration
◦ Working knowledge of Model Context Protocol (MCP) — server design, tool registration, client integration, and security considerations for enterprise contexts.
◦ Experience implementing integration patterns that connect AI platforms to enterprise data sources including CRM, ERP, data warehouse, and document repositories.
◦ Proficiency with REST API design and consumption, event-driven integration patterns, OAuth 2.0, and enterprise identity frameworks applied to AI workloads.
◦ Ability to build extensible MCP server implementations that support multiple use cases without architectural rework.
2. AI Agent Implementation
◦ Hands-on experience building multi-agent systems — task routing, tool-use, memory, orchestration, and failure handling in production or near-production environments.
◦ Familiarity with agent frameworks (LangGraph, AutoGen, CrewAI, or equivalent) and practical judgment about when to use them versus building leaner implementations.
◦ Understanding of agentic failure modes — prompt injection, tool misuse, unbounded execution, runaway cost — and standard mitigations.
◦ Experience implementing human-in-the-loop oversight patterns for AI agents operating within enterprise systems.
◦ Exposure to synthetic workforce or persistent agent identity patterns is strongly preferred.
3. AI-Enabled Delivery Workflow
◦ Experience with AI-augmented software delivery workflows — role-based prompt libraries, structured artifact generation, and LLM-assisted SDLC tooling.
◦ Ability to translate delivery methodology into repeatable tooling: project configurations, prompt libraries, and workflow SOPs that teams can adopt independently.
◦ Understanding of how AI compresses delivery timelines across product management, architecture, engineering, QA, and DevOps functions.
4. Enterprise RAG Pipeline Development
◦ End-to-end RAG pipeline implementation experience: document ingestion, chunking, embedding model selection, vector database configuration, retrieval optimization, and reranking.
◦ Experience defining and measuring retrieval quality metrics — not just working pipelines but pipelines with measurable performance baselines.
◦ Understanding of hybrid search patterns, context window management, and latency versus quality trade-offs at enterprise content scale.
◦ Ability to assess content quality as an input constraint — recognizing when source data richness limits what the AI layer can deliver.
5. LLM Application Development
◦ Production experience building applications on top of LLM APIs including structured output generation, function and tool calling, context management, and multi-turn conversation design.
◦ Strong prompt engineering capability including system prompt design, few-shot patterns, chain-of-thought, and prompt versioning for production systems.
◦ Experience with LLM evaluation frameworks and regression testing for model behavior across prompt and model version changes.
◦ Familiarity with multiple foundation model providers (Anthropic Claude, OpenAI, AWS Bedrock) and the integration implications of each.
6. AI Governance & Data Practices
◦ Experience implementing AI model lifecycle practices: dev/stage/prod promotion criteria, versioning standards, and production readiness gates.
◦ Working knowledge of FinOps tagging and cost attribution patterns for AI workloads including token consumption and inference cost tracking.
◦ Familiarity with AI explainability requirements for enterprise contexts — what documentation is needed, for which workloads, and how to instrument for it.
7. Cloud Architecture & Engineering
◦ Hands-on AWS experience required — Bedrock, Lambda, ECS, S3, RDS, API Gateway, and IAM applied to AI workload design and implementation.
◦ Azure familiarity required — AI Foundry, Container Instances, Cosmos DB, and OpenAI Service as secondary platform context.
◦ Proficiency in Python (required) and Node.js/TypeScript (strongly preferred) for AI application development, pipeline construction, MCP server implementation, and integration scripting.
◦ MLOps experience: CI/CD for AI, Docker, prompt management, model evaluation automation, and observability instrumentation for production AI services.
Soft Skills
• Shifts between architecture contribution, prototype delivery, and embedded team support depending on project phase — often across multiple programs simultaneously.
• Communicates technical designs and trade-offs clearly to both engineering peers and non-technical stakeholders without changing what is true for each audience.
• Produces architecture artifacts that engineers can actually build from — not diagrams that look complete but omit the decisions that matter.
• Brings intellectual honesty to capability assessments — clearly distinguishes what is proven from what is aspirational before commitments are made.
• Works effectively within enterprise governance frameworks, treating ARB processes and documentation requirements as delivery disciplines, not overhead.
• Self-directed with strong judgment about when to align with stakeholders versus when to move with informed autonomy.
Preferred Qualifications
• Education: Bachelor's degree in Computer Science, Software Engineering, or a related technical field. Master's degree preferred.
• Experience: 6–8 years in software engineering, AI/ML engineering, or solution architecture — with a minimum of 2–3 years working directly on large language model or generative AI systems in production.
• Demonstrated history of delivering AI capabilities from proof-of-concept through production deployment with observable outcomes and governance documentation.
• Certifications: AWS Certified Solutions Architect, AWS Certified Machine Learning Specialty, or equivalent cloud and AI credentials.
• Enterprise governance: prior experience contributing to Architecture Review Board processes, FinOps accountability frameworks, or formal change management programs.
• Synthetic workforce or persistent agent platforms: exposure to autonomous agent systems beyond conversational AI — job-scoped agents, tool-use governance, and observable execution — is strongly preferred.
• AI-Enabled Delivery: experience contributing to AI-augmented SDLC workflows that produced measurable delivery compression.
• Industry context: background in live events, trade shows, hospitality, or high-volume B2B enterprise platforms is a differentiator, not a requirement.
Never miss an opportunity! Create an alert based on the job you applied for.
