Agentic AI Architect - Forward Deployment Engineer - Retail

Mackinnon Bruce International; United States; Remote; USD 270,000 - 310,000; Full time

Job summary

Mackinnon Bruce International seeks a Forward Deployment Engineer to lead the technical direction of agentic AI deployments across customer digital platforms. You will own architectures, orchestrate multi-agent workflows, and govern tool access while partnering with stakeholders to deliver production-ready systems.

The role demands hands-on coding, architectural trade-offs, and clear communication with executives. Remote work is supported with preference for West Coast or Texas-based candidates.

Qualifications

• 12–18 years in enterprise/platform architecture.
• 3–4+ years delivering AI/ML or GenAI systems in production.
• Shipped agentic AI systems.
• Production experience with LLM-based agents.

Responsibilities

• Own the technical direction of agentic AI solutions across applications such as loyalty, payments and commerce workflows.
• Define multi-agent orchestration patterns and agent memory across contexts.
• Establish Model Context Protocol (MCP) standards for governed tool access.
• Set LLMOps standards including prompt/versioning and evaluation harnesses.
• Lead implementation reviews and provide hands-on technical leadership.

Skills

FDE experience

Python proficiency

LLMOps

Architectural design

Executive communication

Agent frameworks

ReAct patterns

Multi-agent orchestration

RAG pipelines

Context & prompt engineering

Tools

LangGraph

LangChain

Semantic Kernel

CrewAI

AutoGen

Kafka

LangSmith

Langfuse

Job description

Forward Deployment Engineer | Agentic AI & Retail

Salary: $270,000 - $310,000 + 30% Performance Bonus

Our client is a global leader in data analyitics and AI solutions for businesses across retail and energy sectors, we’re looking for a hands-on Forward Deployment Engineer who has delivered agentic systems into production. The role is remote, with the ideal candidate being based on the west coast, or Texas, though candidates from other states who meet all requirements will be considered.

You’ll act as the technical anchor for an agentic AI initiative across customer-facing digital platforms, translating business objectives into scalable, governed, production‑ready systems.

This role combines engineering, architecture and business-facing delivery. You’ll write code, review implementations, make architectural trade-offs and brief senior stakeholders.

The role

You’ll own the technical direction of agentic AI solutions across applications such as loyalty, payments and commerce workflows, maintaining architectural standards across delivery teams.

End-to-end architecture and orchestration

• Design and govern agentic architectures for customer-facing digital experiences.
• Define multi-agent orchestration patterns, including orchestrator/subagent hierarchies and multi-agent handoffs.
• Architect tool and function calling, stateful graph‑based workflows and agent memory across short‑term, session and long‑term contexts.
• Establish Model Context Protocol (MCP) standards for governed access to tools and enterprise services.

Production readiness and governance

• Define runtime guardrails, including spend controls, fraud detection hooks, human‑in‑the‑loop checkpoints and audit trails.
• Design against agent failure modes such as hallucination, tool misuse, runaway costs and infinite loops.
• Set LLMOps standards covering prompt versioning, prompt/model lifecycle management, RAG pipeline governance and evaluation harnesses.
• Establish observability across latency, token costs, drift and safety.

Retail integration and delivery

• Lead AI agent integration with retail systems, including catalogue, inventory, pricing, promotions, loyalty and checkout.
• Work with API‑first services, distributed systems, event‑driven architectures and real‑time data.
• Address enterprise data quality challenges when connecting agents to operational systems.
• Write reference implementations and lead code reviews, staying directly involved in technical delivery.

Technical leadership

• Act as the primary technical interface between business leaders, delivery teams and platform partners.
• Translate agentic AI concepts into clear business value for executive stakeholders.
• Set architectural boundaries, make trade‑off decisions and defend them under pressure.
• Lead across onshore/offshore delivery models while maintaining a consistent architectural approach.

The non‑negotiables:

• FDE experience

You must have significant experience combining hands‑on engineering with architecture and stakeholder‑facing delivery.

• Enterprise architecture and production AI experience

12–18 years in enterprise/platform architecture, including 3–4+ years delivering AI/ML or GenAI systems in production. You’ll need a strong track record of building and scaling applications.

• Shipped agentic AI systems

You must have production experience with LLM‑based agents. Be ready to explain what you personally built, the architectural decisions you made and how the solution operated in production.

Python proficiency is expected. You must remain capable of writing reference implementations, reviewing code and engaging directly with implementation decisions.

• Agent frameworks and architectural patterns

You’ll need hands‑on experience with frameworks such as LangGraph, LangChain, Semantic Kernel, CrewAI or AutoGen, alongside sound judgement about when and how to use agentic patterns.

Your working knowledge should cover:

• ReAct and plan‑execute patterns.
• Multi-agent handoffs and orchestration.
• RAG pipeline design, including chunking, hybrid retrieval and reranking.
• Context and prompt engineering.
• MCP and governed tool access.
• Agent failure modes and the architectural controls needed to address them.
• LLMOps and observability

Experience with evaluation frameworks, prompt/model lifecycle management and observability tools such as LangSmith, Langfuse or equivalent is needed.

• Enterprise systems and data integration

You’ll need strong foundations in API‑first design, distributed systems, and event‑driven architectures, including Kafka or equivalent, with an understanding of enterprise data quality realities.

• Retail or commerce exposure

Meaningful exposure to retail commerce systems is essential, covering areas such as catalogue, pricing, promotions, loyalty, checkout or payments. Direct retail experience is strongly preferred, but relevant adjacent commerce experience will be considered.

• Leadership and communication

You must be comfortable communicating with executive stakeholders, making architectural decisions under pressure and leading across onshore/offshore teams.

What would be an advantage?

Familiarity with semantic models or knowledge graphs as grounding layers for agents would be beneficial.

Who this will suit

An engineer who can move confidently between code, architecture and business conversations. You’ll be comfortable working in a fast‑moving technology environment with evolving standards, while keeping production reliability, governance and business value central to your decisions.