Closed listing
Senior Machine Learning Engineer
- Toronto, ON, Canada
- Greater Toronto Area, Canada, Canada
Full-time
This opening has closed.
The original listing is kept below for reference.
Job details
- Job type
- Full-time
- Closing date
- Posting language
- English
- Posted
Similar jobs hiring now
Search current openingsOriginal job description
Archived listing. The details below describe a past opening.
Senior AI Engineer, Agentic Systems
Start Date - October 2026
📍 [Toronto / Hybrid] | Full-time
Our client is a global financial services leader making a major strategic investment in AI. They're hiring a Senior AI Engineer to design, build and own the agentic systems that put their models in front of business users, and to make those systems reliable enough to stay there.
What you'll do
- Build agentic systems end-to-end, covering agent loops, orchestration, memory and state, and tool integrations
- Diagnose and fix real-world failures such as stalled loops, tool misuse and behaviour that drifts between runs
- Build evaluation sets and regression checks that reflect real workflows, so quality is measured continuously and not only at demo time
- Balance latency, answer quality and token cost for each workflow
- Connect agents to systems of record and take them from prototype to production
What you'll bring
- Deep expertise in production Python
- Hands-on experience with agent frameworks and protocols such as the Claude Agent SDK, LangGraph, MCP or equivalents
- Experience building and debugging agentic systems in production
- Rigour in evaluation, with measurement designed for business use rather than demos
- Strong engineering discipline: version control, tests, reproducibility and maintainable code
- A background as either a software engineer who moved into AI, or an AI engineer with strong systems fundamentals
Nice to have
- Agent tracing and evaluation tools, such as LangSmith, W&B or custom solutions
- Event-driven or distributed systems, such as Kafka
- Retrieval and memory design for long-running agents
- Full-stack experience integrating agents into user-facing apps
More about this job
Responsibilities
Design, build, and operate production agentic systems, including agent loops, orchestration, memory, state, and integrations with business systems. Diagnose production failures, create workflow-based evaluations and regression checks, optimize latency, answer quality, and token costs, and move agent prototypes into production.
Requirements
Requires deep production Python expertise and hands-on experience building and debugging agentic systems using frameworks or protocols such as Claude Agent SDK, LangGraph, or MCP. Candidates should bring strong systems fundamentals, rigorous evaluation practices, and disciplined engineering practices including version control, testing, reproducibility, and maintainable code.
Skills
- Production Python
- Agentic Systems
- Agent Frameworks
- Claude Agent SDK
- LangGraph
- Model Context Protocol (MCP)
- Agent Orchestration
- Agent Evaluation
- Debugging
- Regression Testing
- Tool Integration
- Version Control
- Reproducible Engineering
- Distributed Systems
- Retrieval and Memory Design
- Full-Stack Development
Visa sponsorship
Not detected in the job text
Categories
- Technology
- Software
- Data & Analytics
- Engineering
- Finance & Accounting
Keywords
- Artificial Intelligence
- Agentic Systems
- Production Python
- Claude Agent SDK
- LangGraph
- Model Context Protocol (MCP)
- Agent Loops
- Orchestration
- Memory and State
- Tool Integrations
- Production Systems
- Evaluation Sets
- Regression Checks
- Latency Optimization
- Answer Quality
- Token Cost
- Systems of Record
- Agent Tracing
- LangSmith
- Weights & Biases
- Event-Driven Systems
- Distributed Systems
- Kafka
- Retrieval Design
- Full-Stack Development
- Version Control
- Testing
- Reproducibility
- Financial Services