Staff Software Engineer - (Linux & GPU Environment) - Substrate Team

Canva is seeking a Staff Software Engineer to join its Substrate Team, focusing on Linux and GPU environments. This role is based in Melbourne, Australia, within Canva's vibrant campus that fosters creativity and collaboration. The Substrate Team is integral to Canva's Core Infrastructure Group, responsible for maintaining and enhancing the platform's Linux fleet, GPU clusters, and compute layers. As Canva continues to scale, this team plays a crucial role in ensuring the stability and efficiency of backend services and AI workloads.

In this position, the Staff Software Engineer will be responsible for several key areas:

- **Linux Orchestration at Scale**: Extending the CAPI interface to support additional providers and optimizing Linux orchestration across various cloud providers and on-premises environments.

- **GPU Clusters**: Managing GPU clusters that handle AI training and inference workloads, including kernel tuning, scheduler decisions, and I/O path optimizations.

- **Bare Metal Infrastructure**: Developing hardware configurations and hypervisor patterns to transform lab environments into production infrastructure, supporting Canva-owned GPUs at data center scale.

- **Linux Internals**: Implementing kernel-level changes, integrating EBPF logic, and performing performance tuning at the system level.

- **Technical Direction**: Establishing best practices for the infrastructure group, authoring design documents, and guiding the evolution of Linux usage within Canva.

- **Hands-On Engineering**: Engaging directly with code, deploying to production, and making impactful changes to enhance backend services and AI workloads.

Ideal candidates will possess experience in several areas:

- **Linux Infrastructure**: Proven track record in building and improving Linux infrastructure at a production scale, with the ability to debug issues across the kernel and application layers.

- **Large-Scale Infrastructure**: Experience operating infrastructure at scale and addressing complex, ambiguous technical challenges.

- **GPU or Specialized Compute**: Background in constructing GPU clusters, AI training infrastructure, or other large compute systems where capacity, cost, and performance are critical considerations.

- **Hardware and Bare Metal**: Familiarity with on-premises infrastructure, hypervisor configuration, and the physical aspects of compute platforms.

- **Ambiguous Technical Problems**: Demonstrated ability to take ownership of unclear problems, define potential solutions, and test the most effective path forward.

- **AI Tools in Engineering Work**: Experience utilizing AI tools in engineering tasks and understanding the implications of AI integration on engineering processes.

While specific compensation details are not provided, Canva offers a comprehensive benefits package that includes:

- **Parental Leave**: Inclusive parental leave entitlements that support all families from day one, regardless of gender or family formation.

- **Equity Packages**: Designed to align individual success with the company's growth, offering equity options to employees.

- **Vibe & Thrive Allowance**: An annual allowance for employees to invest in their wellbeing and professional development.

- **Flex Leave**: Additional paid leave to provide flexibility for personal commitments and needs.

- **Force for Good**: Three days of paid volunteer leave to encourage employees to contribute positively to their communities.

Canva's culture is characterized by a people-first approach, emphasizing inclusivity, flexibility, and a strong sense of community. The company supports various employee resource groups and clubs, fostering connections over shared interests. With a focus on continuous learning and development, Canva provides opportunities for employees to grow both personally and professionally. The hybrid work model offers flexibility, allowing employees to choose between working from home or the office, depending on their preferences and team needs. This environment is designed to empower employees to do their best work and contribute meaningfully to Canva's mission of empowering the world to design.