About the role

Senior AI Engineer designing and building agent and RAG systems for NetBrain's network automation platform. Combining engineering with system-level thinking to enhance reliability and scalability.

Responsibilities

Design and build core agent platform capabilities, including orchestration patterns (e.g., ReAct, Plan-and-Execute), tool execution layers, memory/context management, and guardrails
Develop reusable agent skills and a standardized tool ecosystem aligned with platform conventions
Implement enterprise-grade controls for agent execution, including human-in-the-loop workflows, permissioning, and policy enforcement
Design and implement RAG services, including hybrid retrieval, reranking, embeddings, and citation behavior
Define evaluation frameworks and quality metrics to ensure grounded, reliable outputs
Build automated evaluation and regression pipelines to maintain quality at scale
Establish observability (tracing, metrics, alerting) and optimize systems for latency, throughput, and cost
Define best practices and incident response processes to ensure production readiness
Prototype and productionize emerging approaches (e.g., GraphRAG, knowledge graphs, MCP integrations)
Evaluate AI frameworks and infrastructure (e.g., LangChain, LangGraph, AutoGen, LlamaIndex) and guide platform decisions
Diagnose and resolve complex AI system issues (retrieval quality, hallucinations, prompt regressions, agent failures)
Lead system design and drive technical decisions through hands-on prototyping and benchmarking

BS in Computer Science, AI, Electrical Engineering, or related field (advanced degree preferred or equivalent experience)
5+ years of software engineering experience, including building and deploying production-grade LLM, agent, or RAG systems
Proven track record of shipping and supporting AI-powered features in production
Strong Python skills (API design, testing, error handling, environment management)
Working proficiency in C# or C++
Deep understanding of RAG system design and evaluation methodologies
Experience building evaluation frameworks (test sets, metrics, regression pipelines)
Understanding of LLM security risks (prompt injection, data leakage, agent misuse)
Experience with graph databases, knowledge graphs, or GraphRAG preferred
Experience designing human-in-the-loop systems and execution controls preferred
Experience with agent frameworks (e.g., LangChain, LangGraph, AutoGen, LlamaIndex)
You are willing and able to commute to our Toronto, ON office on a hybrid (once a week) schedule
Manual Dexterity: Repetitive motion of wrists, hands and fingers for using a computer.
Stationary Tasks: Sitting for extended periods, remaining in a stationary position.