AI Engineer; Agentic Workflows

Position Title:

AI Engineer (Agentic Workflows)

Department:
Research & Development

Company:
Protocase Inc.

Location:

Waterloo, Ontario, Canada

Work Arrangement:
In-Office

Employment Type:

Full-Time

Shift: Dayshift

Level: Mid–Senior (Expected

Experience:

3+ Years)

Full Transparency:
We Work a Little Differently Around Here.

This isn't just another job where you punch in and punch out. We come to work every day because we strive to earn a living in a meaningful way. Too many people exist without finding joy or purpose in their work, and we believe that's just wrong. After all, we spend at least a third of our lives working—so why not create workplaces where people want to come to work in the morning and feel good when they leave at the end of the day?

At Protocase, your ideas don't just matter—they shape our future. We thrive on collaboration and innovation, where every voice is heard, valued, and makes a real impact. Does this sound like something you'd want to be a part of and help grow?

Protocase specializes in custom sheet metal enclosures, panels, and parts for engineers and innovators worldwide, built with unmatched speed and quality. We take the impossible and make it possible for some of the brightest minds on the planet.

Learn more about us:

As an AI Engineer (Agentic Workflows) at Protocase, you will design and build applications, automations, and agentic systems on top of existing large language models (LLMs) and foundation models.

This role owns the layer between a model's raw capability and a working business solution: orchestration, retrieval, tool integration, prompt and context design, and evaluation. You'll think in terms of workflows and outcomes—identifying where an agentic or AI-assisted process can replace or augment manual work—and move from abstract problem statements to concrete, working pipelines.

This role does not require deep expertise in training or fine-tuning models from scratch. It requires strong software engineering fundamentals, systems thinking, and fluency in the current landscape of LLM APIs, agent frameworks, and retrieval tooling.

• Design and build agentic workflows that plan, call tools, and chain multi-step actions to complete tasks with minimal human intervention.
• Integrate large language models into products and internal tools via APIs and open-source inference/serving frameworks.
• Architect retrieval-augmented generation (RAG) pipelines, including document ingestion, chunking strategy, embedding selection, and vector database integration.
• Design and iterate on prompts, system instructions, and context-management strategies to reliably steer model behavior.
• Select and integrate appropriate orchestration tooling to coordinate multi-agent or multi-step workflows.
• Build evaluation harnesses and test suites to measure agent and workflow accuracy, reliability, and regression over time.
• Translate ambiguous, high-level business problems into scoped technical specifications and working prototypes.
• Collaborate with product, operations, and subject-matter stakeholders to identify high-value automation and agentic opportunities.
• Monitor deployed AI workflows in production, using observability tooling to track latency, cost, and output quality.
• Stay current with the fast-moving AI agent and tooling ecosystem and recommend adoption of new frameworks, models, or techniques where warranted.

Everything listed in this section is considered required to perform this role successfully.

• Bachelor's degree in Computer Science, Software Engineering, or equivalent practical experience.

• 3+ years of professional software development experience.
• Demonstrated experience building applications on top of LLM APIs, including API providers or self-hosted inference.
• Practical experience with at least one agent orchestration framework or workflow automation platform.

• Solid understanding of RAG architecture: embeddings, vector search, chunking, and retrieval strategy.
• Strong general-purpose programming ability in Python and/or a modern backend language.
• Ability to reason abstractly about a business problem and…