Staff AI Developer (AI)

Location: Toronto, ON, or Canada remote; mandatory weekly on-site client delivery 3 days per week in the United States. Candidates must be legally authorized to work in Canada and eligible to secure immediate U.S. non-immigrant work authorization, such as TN status under USMCA.

Salary: $165,604–$228,491 USD annually.

Company

Robots & Pencils designs and delivers production-ready AI systems that integrate into enterprise operations and real-world workflows.

What you will do

• Design and build agentic workflows with reasoning loops, tool and function calling, and single- or multi-agent orchestration.
• Build and maintain RAG pipelines covering chunking, embeddings, vector search, re-ranking, and knowledge-source refresh handling.
• Integrate AWS Bedrock, Agent Core, MCP-based tools, and production system prompts.
• Develop evaluation and observability systems using golden datasets, RAGAS-style metrics, LLM-as-judge, and tracing tools.
• Build Python and Node.js backend services, serverless AWS architectures, DynamoDB schemas, and event-driven workflows.
• Lead architecture decisions, mentor engineers, and own features and releases through debugging, hardening, deployment, and production troubleshooting.

Requirements

• 6+ years of professional software engineering experience, including meaningful experience shipping GenAI or LLM-powered systems in production.
• Hands-on expertise in agentic AI, including reasoning loops, tool or function calling, multi-agent orchestration, and evaluating single-agent versus multi-agent designs.
• Practical RAG expertise with chunking strategies, embeddings, vector databases such as OpenSearch, cosine similarity search, and re-ranking.
• Experience building LLM evaluation and observability with golden datasets, LLM-as-judge, RAGAS or comparable metrics, and LangFuse, LangSmith, or equivalent tracing.
• Strong prompt engineering skills, including structured production system prompts, output constraints, and few-shot design.
• Hands-on experience with AWS Bedrock, Agent Core, Lambda, DynamoDB single-table design, S3, SQS, EventBridge, and Step Functions.
• Strong Python and Node.js skills, including full-stack application and RESTful API development.
• Understanding of production reliability patterns for LLM systems, including retries with backoff, circuit breakers, and fallback models.
• Experience with Docker, cloud-native deployment, ambiguous integration-heavy problems, and end-to-end ownership.
• Legal authorization to work in Canada and eligibility for immediate U.S. non-immigrant work authorization for mandatory weekly U.S. on-site delivery.

Nice to have

• Direct Amazon Bedrock Agent Core experience, including agent registration, MCP tool wiring, memory and session management, and gateway provisioning.
• Experience in regulated or high-stakes domains such as education, healthcare, or financial services.
• Familiarity with Step Functions, Sequencer, or similar state-machine workflow tools.
• Production experience with LLM observability tooling.

Culture & Benefits

• Work on cross-functional teams delivering production-ready AI systems in 30–45 days.
• Collaboration with product, design, and delivery leads across global teams.
• Paid time off and medical, dental, and vision insurance.
• 401(k) benefits for eligible employees.
• Employment may be conditional on a background check under local legislation.