Location

Toronto, ON (Hybrid - x3 per week in downtown Toronto office)

Type

Full-Time

We're partnering with a highly technical AI organization building the infrastructure that powers production AI systems at massive scale. Operating as a AI innovation startup within a much larger global technology business, the team combines the pace, ownership, and greenfield engineering opportunities of an early-stage company with the resources and reach of an established platform serving hundreds of millions of users.

This is not an AI research role. We're looking for a Staff-level software engineer who can define the architecture and technical direction for large-scale AI infrastructure across inference, agentic systems, distributed platforms, and cloud-native backend services.

You’ll help build the infrastructure used to deploy and operate text, voice, vision, code, and domain-specific models, while architecting the runtime, orchestration, safety, and developer tooling required for autonomous AI agents to operate reliably in production.

This is a highly hands-on Staff position. You’ll solve complex engineering problems, lead major technical initiatives, establish platform standards, and influence how production AI systems are built across the organization.

What You'll Do

Define the technical architecture and direction for production AI platforms across inference and agentic systems

Architect and build the runtime, orchestration, and developer tooling required for autonomous AI agents

Design multi-agent coordination systems that enable agents to reason, collaborate, use tools, and execute complex workflows

Build multi-model serving infrastructure across text, voice, code, vision, and domain-specific models

Own the complete model lifecycle, including deployment, serving, monitoring, updating, routing, and model swapping

Optimize inference performance across latency, throughput, reliability, and cost using batching, caching, quantization, and intelligent routing

Build secure tool-use infrastructure that allows agents to interact safely with APIs, databases, and internal services

Develop guardrails covering permissioning, sandboxing,



prompt injection, data leakage, and human-in-the-loop oversight

Build evaluation, observability, and monitoring frameworks that measure agent behaviour, detect regressions, and diagnose non-deterministic failures

Design scalable backend services, APIs, event-driven systems, and durable workflows supporting production AI applications

Develop SDKs, APIs, and platform capabilities that allow internal teams to build and deploy AI agents quickly and safely

Lead complex, cross-functional technical initiatives in partnership with Product, ML, Infrastructure, and Security teams

Establish engineering standards and best practices for agent design, model serving, tool calling, evaluation, and production reliability

Mentor senior engineers and raise the technical bar across the broader engineering organization

What We're Looking For

8+ years of software engineering experience, including significant experience building large-scale backend or distributed systems

3+ years of experience building production AI systems, LLM applications, agentic platforms, or machine learning infrastructure

Demonstrated experience owning the architecture and delivery of complex, business-critical technical initiatives

Strong understanding of LLM-based agent architectures, including tool use, memory, planning, multi-step workflows, and multi-agent coordination

Experience building highly reliable distributed systems using event-driven architectures, task queues, state management, and durable workflows

Experience evaluating production LLM systems, building automated evaluations, detecting regressions, and debugging non-deterministic failures

Strong programming experience with Python and/or TypeScript, with the ability and willingness to work across both





Experience with Kubernetes, Docker, AWS or GCP, and modern cloud-native deployment practices

Experience working with commercial LLM APIs, open-source models, or model-serving technologies

Understanding of inference optimization techniques such as quantization, batching, caching, routing, and GPU utilization

Strong understanding of the security risks associated with agentic systems, including prompt injection, privilege escalation, and data leakage

Exceptional system-design and software-engineering fundamentals

Strong written and verbal communication skills, with the ability to influence technical direction across teams

Comfortable operating in an ambiguous, fast-moving workplace with substantial ownership and autonomy

Passion for building production software and infrastructure rather than purely research-focused AI

Nice to Have

Experience with model-serving technologies such as vLLM, TensorRT-LLM, or Triton

Experience with Temporal, Airflow, Prefect, or similar workflow-orchestration platforms

Familiarity with Model Context Protocol (MCP) or other agent communication standards

Experience with model fine-tuning, LoRA, or quantization

Experience building AI infrastructure within fintech, healthcare, or another regulated industry

Experience working with multimodal, voice, or edge-inference systems

Experience designing human-in-the-loop approval and oversight systems

Experience building developer platforms, internal SDKs, or CI/CD automation for AI workloads

Why Apply?

Take Staff-level ownership over the architecture and technical direction of major AI platforms

Build inference and agentic infrastructure used across a global technology organization

Work on genuinely greenfield engineering problems spanning LLMs, autonomous agents, distributed systems, security, and cloud infrastructure

Remain deeply hands-on while influencing engineering standards and mentoring a high-calibre technical team

Join a startup-style environment with significant autonomy, backed by the scale and resources of an established global platform

📌 Staff AI Engineer (Toronto)
🏢 Arcadia
📍 Toronto