As a Senior Machine Learning Engineer in the Integrations and Solutions Group at Canva, you will shape how AI platforms interact with Canva by building and optimizing AI-ready tools, APIs, and agent architectures. You'll enable seamless, reliable, and scalable integrations with leading AI assistants while driving measurable improvements through evaluation and observability systems.
What You'll Do
- Design and optimise AI-ready tools and APIs that enable LLM platforms (ChatGPT, Claude, Gemini and others) to reliably interact with Canva's design capabilities — focusing on optimising tools to decrease latency and reduce costs without sacrificing end-user performance.
- Pioneer agent to agent communication.
- Build and maintain evaluation frameworks to systematically measure tool-use accuracy across platforms, including proxy-based fast evals for development velocity and real-client evals for production validation.
- Experiment with LLM orchestration and agent architectures — Develop Canva agents that any 3rd party provider can call to design quickly, efficiently and at scale.
- Automate entire workflows such as marketing campaigns by connecting user intent, Canva and third parties.
- Develop observability and monitoring systems to track how AI assistants consume Canva's tools in production, identifying failure patterns and driving continuous improvement.
- Work across platform boundaries — understanding the behavioural quirks of different LLM platforms and building adapters or configurations that account for them at scale.
- Collaborate with AI partners and internal teams to shape tool design, API evolution, and integration strategy aligned with Canva's broader AI ecosystem goals.
What We're Looking For
- You have experience working with LLMs in production — not just prompt engineering, but rigorous error analysis and understanding how models select and invoke tools, where they fail, and how to improve reliability.
- You're skilled in evaluation and benchmarking — you know how to design metrics, build eval pipelines, and measure whether AI integrations are actually working at scale.
- You have experience designing APIs or tools consumed by AI agents — understanding that tool descriptions are UX for AI, and small changes in naming or structure can swing accuracy by 40%.
- You can build and iterate on agent architectures — whether that's orchestration layers, MCP servers, or intent-based APIs that abstract complexity from external models.
- You're proficient in Python and familiar with ML frameworks, but equally comfortable working with TypeScript/Node.js and cloud infrastructure (Cloudflare Workers, AWS, or similar).
- You thrive in cross-functional, partner-facing work — collaborating with engineering, product, and external AI platform teams (OpenAI, Anthropic, Google) to shape integration strategy.
- You enjoy solving ambiguous, fast-moving problems — the AI ecosystem is evolving rapidly and you're energised rather than overwhelmed by that.
Technical Stack
- Python
- TypeScript
- Node.js
- Cloudflare Workers
- AWS
- ML frameworks
- LLMs
- MCP platform
- REST API
- Apps SDK
- AI agent tools
- evaluation frameworks
- observability systems
- agent architectures
- LLM orchestration
- APIs
Team & Environment
- Part of the Ecosystem team, specifically the Integrations and Solutions Group, which builds Canva's AI integration layer and works directly with external AI partners.
Work Mode
- Remote within Australia
Canva is committed to creating a diverse and inclusive environment and is proud to be an equal opportunity employer. We celebrate all forms of diversity and are dedicated to fostering an equitable workplace for everyone.
