Developer, Agentic Systems & Automation

Orium

About The Position

In 2026, Orium is investing in AI and automation to improve how SG&A functions operate across the organization. Our goal is to build agentic systems that support real human decision-making — including a unified entry point and two-way workflows that allow agents, humans, and systems to interact reliably. We’re approaching this work intentionally, with strong foundations around trust, escalation, and human judgment. We’ve completed early experimentation through pilots and demos, and we’re now looking for someone to help turn that learning into durable, production-ready capabilities. This is a contract role focused on hands-on execution with strong technical ownership. As a Developer, Agentic Systems & Automation, you’ll work hands-on to design and build the technical foundations of these systems, partnering closely with internal stakeholders to translate evolving needs into reliable, scalable agentic workflows. The focus of this engagement is to accelerate the build-out of core agentic capabilities and to establish strong technical foundations over a defined period.

Requirements

Strong proficiency in JavaScript/Typescript, including building and deploying backend services, integrations, or extensions to support workflow and automation systems.
Strong experience building backend systems, workflows, or integrations, including APIs, event-driven systems, and automation frameworks.
Comfortable working with orchestration tools and workflow engines (e.g., Zapier + code, n8n, Temporal, Airflow, or similar), including extending platforms with custom logic where needed.
Experience building, integrating, or supporting AI / LLM-powered systems, with an emphasis on reliability, validation, and safe failure.
Experience designing or contributing to system architecture, particularly in integration-heavy or distributed systems.
Thoughtful approach to system design, trade-offs, and long-term maintainability.
Comfortable taking ownership in ambiguous problem spaces, helping define the path forward rather than waiting for fully defined requirements.
Strong collaborator who enjoys building in partnership with product, domain, and systems owners.

Responsibilities

Partner in shaping the technical approach and architecture of agentic systems, helping translate evolving needs into practical, scalable solutions.
Design and implement agentic workflows that support two-way interaction between humans, agents, and internal systems.
Build and orchestrate workflow logic using tools and frameworks such as Zapier + code, n8n (or similar), APIs, and event-driven architectures.
Partner to integrate people-focused agentic workflows into a cohesive, reliable entry-point experience.
Bring a technical perspective to early-stage ideas and prototypes, helping assess feasibility, trade-offs, and paths to production.
Help evolve early implementations into more robust, maintainable, and production-ready systems.
Help establish and document patterns for reliability, observability, and safe failure in agentic workflows.
Ensure implementations align with defined trust models, escalation rules, and human-in-the-loop principles.
Work hands-on with LLMs, prompts, and agent behaviour to prioritize correctness, clarity, and human accountability.
Evaluate and integrate tools, frameworks, and platforms that support agentic patterns as the system matures.
Contribute to and document shared standards and patterns for AI-augmented systems at Orium.
Design and evolve retrieval and indexing approaches (e.g., RAG, structured knowledge sources) to ensure agents operate on accurate, relevant, and up-to-date context.