Senior Staff Engineer: Agent Infrastructure at Scale

PromptCube; Toronto; Hybrid; CAD 340,000 - 509,000; Full time

Benefits offered by this job

Lunch stipend

Health & dental benefits

RRSP matching

Parental leave top-up

Enrichment benefits

Conferences stipend

Vacation: 6 weeks

Home office stipend

Job summary

Cohere is building the next generation of agentic AI infrastructure in North America, focusing on secure code execution, identity and authentication for AI agents, and scalable model routing across diverse environments. The role emphasizes production-ready ML systems, developer experience, and collaboration with research teams.

The position supports North American work authorization with hybrid attendance at offices in major cities.

Qualifications

• Experience building production ML infrastructure with strong systems fundamentals.
• Hands-on work with agentic systems and multi-agent workflows.
• Familiarity with model routing and LLM provider frameworks across environments.

Responsibilities

• Build secure execution environments for agent-generated code
• Design identity and authentication systems for AI agents
• Develop model routing and orchestration across model types and environments
• Implement rate limiting, quotas, and resource management for agent workflows
• Create state management, memory, and filesystem abstractions for agents
• Prototype and evaluate new technologies for production infrastructure
• Partner with research teams to shape infrastructure for future agent systems

Skills

Production ML infra

Agentic systems

Model routing

Kubernetes

Distributed systems

Tools

Kubernetes

Job description

Cohere is building the next generation of agentic AI infrastructure in North America, focusing on secure code execution, identity and authentication for AI agents, and scalable model routing across diverse environments. The role emphasizes production-ready ML systems, developer experience, and collaboration with research teams.

The position supports North American work authorization with hybrid attendance at offices in major cities.