Software Engineer: Agent Infrastructure

AgentField.ai; Toronto; On-site; CAD 130,000 - 190,000; Full time

Job summary

AgentField.ai in Toronto is seeking a Software Engineer to architect and implement the agent infrastructure for large-scale deployments. You will design execution and scheduling for long-running workloads, build deployment paths including on-prem and air-gapped environments, and contribute to open-source guidance and tooling.

The role requires 5+ years in production backend and distributed systems, hands-on experience with Terraform, Docker, and Kubernetes, and a track record of scalable backend

Qualifications

• Bachelor's degree in Computer Science, Computer Engineering, or a closely related field, or equivalent practical experience.
• At least 5 years of experience building production backend and distributed systems, with previous experience in product-based startup environments.
• At least 3 years of experience contributing to enterprise open-source infrastructure: software shipped to external users and operated by people outside your own team.
• At least 3 years of experience in large-scale distributed compute: job schedulers and workflow orchestration engines; on-prem cluster management and deployment; distributed execution frameworks handling long-running, stateful, resumable workloads.
• Hands-on experience deploying AI agent systems into enterprise production, in a paid professional role at a real enterprise, not prototypes, demos, or side projects. You have dealt with what breaks when autonomous workloads meet real security review, real data boundaries, and real on-call.
• Experience with Terraform, containerization with Docker and Kubernetes, and systems design.
• Proven track record of designing and implementing highly scalable and reliable backend systems.
• Strong analytical and problem-solving skills.

Responsibilities

• Design and build execution and scheduling for long-running agent workloads across heterogeneous compute and sizes.
• Horizontal scaling in various dimensions: from million-agent scale to million identities.
• Build and maintain deployment paths our enterprise users depend on, including on-prem, air-gapped installs, and bring-your-own-cloud.
• Benchmark models across agentic workloads and build advanced agentic applications that demonstrate what the infrastructure makes possible.
• Continuously improve performance and reliability of the runtime through profiling, experimentation, and load testing.
• Participate in open-source stewardship: RFCs, design documents, code review, and the public issue tracker.
• Collaborate with cross-functional teams and directly with enterprise users deploying agents in production.

Skills

Backend development

Distributed systems

Open-source contributions

Education

Bachelor's degree in CS/Engineering or related field

Tools

Terraform

Docker

Kubernetes

Job description

Software Engineer: Agent Infrastructure

Backend / Distributed Systems · Toronto, Canada · Full-time

Design and build the execution fabric, scheduling, and deployment paths that power large-scale agent systems: from million-agent runs to on-prem and air-gapped enterprise installs.

About the role

As a Software Engineer at AgentField, you will build the framework and various pieces around it. You will also benchmark models and build advanced agentic applications on top of AgentField that might go beyond 100k+ agents each run.

Responsibilities

• Design and build execution and scheduling for long-running agent workloads across heterogeneous compute and sizes.
• Horizontal scaling in various dimensions: from million-agent scale to million identities.
• Build and maintain deployment paths our enterprise users depend on, including on-prem, air-gapped installs, and bring-your-own-cloud.
• Benchmark models across agentic workloads and build advanced agentic applications that demonstrate what the infrastructure makes possible.
• Continuously improve performance and reliability of the runtime through profiling, experimentation, and load testing.
• Participate in open-source stewardship: RFCs, design documents, code review, and the public issue tracker.
• Collaborate with cross-functional teams and directly with enterprise users deploying agents in production.

Requirements

• Bachelor's degree in Computer Science, Computer Engineering, or a closely related field, or equivalent practical experience.
• At least 5 years of experience building production backend and distributed systems, with previous experience in product-based startup environments.
• At least 3 years of experience contributing to enterprise open-source infrastructure: software shipped to external users and operated by people outside your own team.
• At least 3 years of experience in large-scale distributed compute: job schedulers and workflow orchestration engines; on-prem cluster management and deployment; distributed execution frameworks handling long-running, stateful, resumable workloads.
• Hands-on experience deploying AI agent systems into enterprise production, in a paid professional role at a real enterprise, not prototypes, demos, or side projects. You have dealt with what breaks when autonomous workloads meet real security review, real data boundaries, and real on-call.
• Experience with Terraform, containerization with Docker and Kubernetes, and systems design.
• Proven track record of designing and implementing highly scalable and reliable backend systems.
• Strong analytical and problem-solving skills.

Bonus points

• Critical contribution to open-source packages, or developing and maintaining open-source packages with outside contributors.

About AgentField

AgentField is a Toronto-based startup building Dev-first infrastructure for AI agent systems. Open source, Apache 2.0. It gives agents a control plane with APIs, distributed execution, state, resilience, observability, identity, governance, and proof: the backbone of large-scale autonomous software.

Learn more at agentfield.ai, our docs, or GitHub.