AI Architect

Kinaxis Inc.

2d•Hybrid

About The Position

We are looking for an AI Enterprise Architect to help shape how AI‑enabled and agent‑driven capabilities are designed, governed, and evolved across Corporate Kinaxis. This is a senior architecture role focused on defining enterprise‑level AI architecture, establishing clear guardrails, and working closely with other architects and risk partners to ensure AI accelerates the business without creating unmanaged risk or long‑term architectural debt. You will operate within a federated Enterprise Architecture model, partnering with Enterprise Architects, Data Architects, Security Architects, Solution Architects, and business stakeholders. Success in this role depends as much on judgment, collaboration, and influence as it does on technical depth

Requirements

8–10+ years of experience in enterprise architecture, solution architecture, or senior technical roles within complex enterprise environments.
Strong grounding in enterprise architecture practices, including reference architectures, integration patterns, and architectural governance.
Sufficient AI architecture literacy to evaluate designs involving AI‑enabled or agent‑based systems, including implications for decision‑making, data access, cost, and operability.
Experience working in a federated architecture model, partnering with other architects and risk stakeholders.
Ability to communicate complex architectural trade‑offs clearly to both technical and non‑technical audiences.

Nice To Haves

Experience contributing to or defining AI‑related architectural standards or guardrails.
Familiarity with enterprise data and integration architectures that support AI workloads.
Exposure to environments where technology is evolving rapidly and ambiguity is the norm.

Responsibilities

Build the “paved road” for enterprise AI: publish and maintain reusable reference architectures, guardrails, and starter patterns that teams can adopt quickly (e.g., standard ways to ground AI in enterprise knowledge, call tools safely, and integrate with systems of record), so safe delivery is the default rather than a special case.
Define and operationalize agent interaction and authority boundaries: help teams design AI workflows where components may collaborate (including multi-step or multi-component decision flows) by establishing clear interaction models, responsibility boundaries, and escalation/override points that are enforceable by design.
Embed governance into platforms and delivery workflows (not paperwork): partner with other architects to translate risk appetite into practical architectural constraints, and ensure those constraints show up in platform choices, standard patterns, and intake/review triggers—so governance speeds teams up instead of slowing them down.
Make AI systems measurable and production-ready: define what “good” looks like for AI-enabled and agentic solutions (quality, reliability, cost, latency, safety), and ensure teams implement evaluation and monitoring approaches that detect drift, failure modes, and cost surprises before they become enterprise incidents.
Create decision-ready architecture artefacts teams can execute: produce practical outputs (architecture decision records, lightweight review checklists, pattern catalogs, and example designs) that reduce repeat debates and let delivery teams move fast with clarity.
Continuously evolve the guardrails as AI changes: maintain an enterprise AI “technology and pattern radar” focused on architectural implications (not hype), and update approved patterns and constraints on a cadence that matches real AI change while preserving stable principles so teams aren’t constantly re-architecting.
Support adoption with hands-on architectural guidance: engage early with teams to shape designs, anticipate integration/data boundary issues, and reduce rework; providing direct guidance grounded in real-world implementation and operational considerations.