Applied AI Engineer (Agentic Workflows & RAG) - Master-Level Internship

Stie Yai; Canada; Hybrid; CAD 90,000 - 120,000; Full time

Job summary

Stie Yai is seeking a highly capable AI Engineer who has recently completed a Master’s in CS/AI/IS. You will design agentic workflows, build RAG pipelines, and select models to balance latency, cost and quality.

Collaborate with Builders and Integration & Data Engineers, document prompts and integrations in Notion, and ensure AI outputs are reliable and safe.

Qualifications

• Currently enrolled in or recently graduated from a Master’s program in CS/SE/AI/IS or related field.
• Hands-on experience with AI coding assistants and Claude API, supported by portfolio or examples.
• Working understanding of modern LLMs and context windows.
• Experience building agentic workflows, RAG/retrieval pipelines, or integrating models via tool calling or MCP.
• Excellent prompt-engineering and context-management skills.
• Coding fluency in JavaScript/React and/or Python to build, evaluate, and fix AI outputs.
• Strong verbal and written communication skills within cross-functional teams.

Responsibilities

• Design, build, and harden agentic workflows for reliable planning and actions.
• Build RAG and search pipelines to ground model outputs in core applications.
• Select models balancing latency, cost, and quality.
• Engineer prompts and context structures for model classes.
• Write evaluations to verify AI features perform consistently.
• Connect AI tools to internal systems via APIs or MCP.
• Review AI-generated code and workflows for correctness, safety.
• Collaborate with Builders and Integration & Data Engineers; document workflows in Notion.

Skills

Prompt engineering

Context management

AI coding assistants・Claude API

Model selection

Cross‑functional collaboration

JavaScript/Python

Communication skills

Education

Master’s degree in CS/SE/AI/IS

Tools

Claude API

Model Context Protocol (MCP)

Tool calling

APIs integration

Job description

Your Role

Here’s what you will be doing:

• Design, build, and harden agentic workflows that plan and take actions reliably, understanding failure modes and structuring tasks for success.
• Build retrieval-augmented generation (RAG) and search pipelines that fetch relevant client data and ground model outputs within core applications.
• Select appropriate models for each task, balancing latency, cost, and quality trade-offs.
• Engineer prompts and context structures tailored to model classes, including framing reasoning models effectively.
• Write evaluations for AI features to verify consistent performance beyond one-time success.
• Connect AI tools to internal systems and data sources via APIs or Model Context Protocol (MCP) to support real client use cases.
• Review and validate AI-generated code and automated workflows for correctness, security, and safety.
• Collaborate with Builders and Integration & Data Engineers to deliver complete solutions and document workflows, prompts, and integrations in Notion.

About You

The company is looking for:

• Current enrollment or recent graduation from a Master’s program in Computer Science, Software Engineering, AI/ML, Information Systems, or related fields.
• Strong hands-on experience with AI coding assistants and Claude API or comparable model APIs, supported by portfolio or examples.
• Working understanding of modern large language models (LLMs) and reasoning models, including context windows and model selection criteria.
• Practical experience in at least one area: building agentic workflows, RAG/retrieval pipelines, or integrating models via tool calling or MCP.
• Excellent prompt-engineering and context-management skills.
• Coding fluency in JavaScript/React and/or Python to build, evaluate, and fix AI-generated outputs.
• An instinct for evaluation to measure AI correctness beyond plausibility.
• Strong verbal and written communication skills within cross-functional teams.

New graduates are encouraged to apply. The company values curiosity, passion for growth, and the desire to create impact, even if all qualifications are not perfectly met.