AI Platform Engineer

Description

AI Platform Engineer

Job Details

Job Type

Full-time, Remote, Occasional travel

IL

• Agentic Systems

Description

HALO Branded Solutions is a global leader in branded merchandise, uniform programs, and recognition solutions, and we're building something new. We're assembling a dedicated Agentic Systems team to design and build the intelligent systems that will transform how we take an order from ideation to delivery. That means reimagining the full order management lifecycle: quote capture, supplier and decorator coordination, shipping, PO and invoicing as a platform where AI agents do the work that today requires dozens of manual handoffs across customers, sales, operations, and finance.

We're not experimenting with AI. We're building production systems that will run the business.

The AI Platform Engineer makes the platform real: building, running, and hardening the AI harness (runtime layer), control plane (governance layer), and operations control center (observability layer) in close partnership with the AI Platform Architect.

This is the role that sits between architectural vision and working production infrastructure, implementing the orchestration framework, building the shared state and memory layer, establishing the integration contracts, and ensuring the platform is reliable enough that every agent running on it actually works. The Architect draws the blueprint; you build the building and keep it standing.

This is not a junior role with a senior title. This is a systems-first engineering role with serious development depth. The decisions you make at the implementation level, how state is managed across agent handoffs, how the runtime is structured so deploying a new agent doesn't require rebuilding the platform, how failure is isolated so one agent's problem doesn't cascade, determine whether HALO's agents work in production. That's the responsibility this role carries.

If you're a strong systems engineer with deep distributed systems instincts who is ready to move into AI infrastructure, or an experienced AI platform engineer who wants to build something from the ground up, this is the role.

Responsibilities

Build and operate the orchestration infrastructure.

Implement the multi-agent orchestration layer that all HALO agents run on, managing routing between agents, retry logic, circuit breakers, payload validation, and sequencing. This infrastructure lives outside of individual agents, not inside them. Framework selection is an Architect-level decision; implementation and ongoing operation are yours.

Own shared state and memory.

Design and implement the state management layer that lets agents pass context across handoffs without losing information: session state, cross-agent memory, and the persistence model that supports both short-lived task agents and longer-running workflow agents.

Build the agent runtime environment.

Own the infrastructure agents execute on: deployment pipelines, environment management, model version pinning, and the scaffolding that makes deploying a new agent a configuration exercise, not an infrastructure project. The Agent Developers should be able to ship a new agent without filing a ticket to you first.

Define and enforce integration contracts.

Work directly with the AI Integration Engineer to establish and maintain the contracts governing how agents request data: format, latency guarantees, error handling. Own the API gateway layer between the orchestration platform and the external systems agents interact with.

Own platform reliability.

Uptime, failure isolation, incident response for agent failures, and the runbook for what happens when an agent behaves unexpectedly in production. This is an ongoing operational responsibility. The platform is never "done" in a way that lets you stop thinking about its reliability.

Provide the observability infrastructure.

Implement the platform-level hooks: logging interfaces, trace endpoints, policy enforcement points that the Governance Engineer instruments. The observability standards live with the Governance Engineer; the infrastructure that makes those standards enforceable lives with you.

Requirements

The team you're joining

You'll work directly with the AI Platform Architect (your technical lead and the source of the architectural decisions you implement), the AI Integration Engineer (your closest peer, jointly owning the contract between the orchestration and data layers), the AI Agent Developers (who build on the platform you operate), and the AI Governance & Observability Engineer (who instruments the hooks you provide and dashboards to observe). It's a small, senior, and deliberately collaborative team building infrastructure that doesn't yet exist at HALO.Requirements

• 5+ years of software or platform engineering with genuine depth in distributed systems: state management, failure modes, event-driven architecture, or API platform engineering
• Experience building or operating infrastructure that other engineers build on, platforms with downstream consumers who depend on their reliability
• Production experience with at least one agent or LLM orchestration framework (AWS AgentCore, LangGraph, CrewAI, LangChain, or equivalent) or strong distributed systems instincts and demonstrated ability to ramp quickly
• Comfort with operational ownership and being accountable for reliability

You'll stand out if you have:

• Hands-on experience building multi-agent orchestration infrastructure in production
• Familiarity with agent state and memory patterns: how context persists across agent turns, sessions, and workflow stages
• Experience with AI observability tooling (LangSmith, Arize, Weights & Biases, or equivalent) from the infrastructure side
• Background in API gateway design and integration contract management between platform layers
• Experience in order management, commerce, or supply chain domains