Staff Software Engineer Data Platform Workflow

Data Platform Workflow Engineer

About

Join the team redefining how the world experiences design. We are looking for someone who calls Melbourne home. Our hybrid way of working gives you the flexibility to work remotely, and to come together on campus for meaningful in-person collaboration and connection when it matters most.

The Data Platform Workflow team owns the data platform users’ analytics journey at Canva. The strategic bet is converting that workflow to AI-friendly or AI-native. The speed at which analytics arrives is the speed at which Canva learns. AI generates models faster than humans can write them, surfaces insight earlier in the journey, and changes what review even means. The team has shipped early proof of what this looks like, and we’ll need to take it to the next levels.

Responsibilities

• Map the end-to-end path a data platform user takes from question to insights, finding the real bottlenecks, anchoring the team’s roadmap in data rather than intuition.
• Build agent-ready platforms by converting command-line and human-first interfaces into MCPs, skills, connectors and APIs that AI agents can use natively, including BI tooling, modelling layers, and lineage.
• Set the standard for what AI-native means inside the data platform, what good looks like, and what other Canva platform teams should be building toward, especially to ensure well governed and reliable AI capabilities at scale.
• Build the metrics that show value: adoption, time saved, bottlenecks removed, so the work compounds rather than getting lost.

Requirements

• Deep familiarity with AI-native platforms: Worked closely with MCPs, agents, orchestration layers, skills, or context engineering systems in production.
• Workflow ownership across teams: Owned an end-to-end developer or analyst workflow at scale, understanding where time actually leaks in a multi-step journey and what to optimise first.
• Data platform background: Worked on the engineering side of a data platform, with knowledge of dbt, lineage, modelling layers and BI tools from the inside.
• Cross-team direction setting: Set technical standards or patterns that engineers in adjacent teams adopted by default. Comfortable holding the long-term architecture and the immediate deliverable at the same time.
• Visionary plus pragmatic: Translated an unsolved problem into a written plan well enough that other engineers could execute against it. Can sit with ambiguity without freezing.
• Adoption mindset: Shipped a platform feature and measured whether anyone actually used it. Treats adoption as part of the engineering work, not a marketing afterthought.
• AI agents and orchestration: Deep practical experience with MCPs, agent skills frameworks, LLM-based orchestration, and their failure modes. Production architectural at scale understanding.
• Languages: Strong in Python for platform and tooling work. Comfortable in Go or Java for adjacent systems.
• Data platform stack: Snowflake, dbt, lineage tooling, modern BI tools. Familiar with CI/CD for data pipelines.
• API and connector design: Can build interfaces that humans and agents both use well.
• CS fundamentals: Concurrency, system design, failure handling, observability. Strong on architecture and design patterns.
• Cloud primary: AWS depth required (GCP a bonus). Understands the cost and reliability trade-offs of managed services at scale.
• AI-assisted engineering: Daily user of Copilot, Claude Code, or equivalent at the architectural level.