Software Developer (Agentic Evaluation)

Autodesk•Vancouver, BC

3d•CA$88,000 - CA$128,700

About The Position

As a Software Developer on the Fusion platform services team within Product Development and Manufacturing Solutions (PDMS), you'll be part of a team of technologists dedicated to creating cutting-edge AI and generative AI solutions that enhance developer productivity and experience. You'll work closely with AI engineers, software architects, and product engineering teams to build and rigorously evaluate intelligent agentic systems — including benchmarking AI agents against commercial solvers — and develop MCP (Model Context Protocol)-based tooling that integrates seamlessly with IDEs such as VS Code and Cursor.

Requirements

BS/MS in Computer Science, Machine Learning, or a related applied AI field
Expertise in Python and ML frameworks (PyTorch, Transformers, scikit-learn)
Experience with Large Language Models applied to software understanding or test generation
Knowledge of AI evaluation methodologies and metrics for agentic task completion and test quality
Strong foundation in statistical analysis and experimental design
Experience with developer workflow and productivity measurement frameworks

Nice To Haves

Background in software engineering or QA with close collaboration with development teams
Familiarity with test automation frameworks (e.g., Playwright, Selenium, Pytest, Appium) and CI/CD pipelines
Experience designing benchmarks that compare AI agents against commercial or domain-specific solvers
Hands-on experience with MCP (Model Context Protocol), building, evaluating, and optimizing MCP servers and tool integrations within agentic pipelines
Experience with agentic AI frameworks including LangGraph, AutoGen, or the Anthropic Agent SDK / Claude Code
Knowledge in vision-language models or multi-modal AI for UI and system-level understanding and evaluation
Experience with Azure AI Foundry/ML or AWS cloud ML platforms

Responsibilities

Develop and orchestrate multi-agent AI systems for automated test generation, test execution, and end-to-end development workflow optimization using frameworks like LangGraph, AutoGen, or the Anthropic Agent SDK (Claude Code)
Design and implement agentic workflows that coordinate multiple AI agents to autonomously drive test automation across UI, API, integration, and system levels, from test case synthesis to result evaluation, ensuring seamless integration with existing developer tools and MCP-compatible services
Build evaluation frameworks and custom benchmarks for agentic systems, including comparisons of AI agents against commercial solvers, using tools like AgentBench and Langfuse
Evaluate MCP server and tool performance across agentic pipelines, measuring latency, accuracy, context fidelity, and end-to-end task completion rates