Principal ML Platform Engineer
Synthesia · Remote · Europe ·
COMPENSATION
€•• – ••k
Listed on the employer's job board (Ashby)
- Principal level role
- Fully remote position
- Python + Kubernetes
- ML infrastructure focus
- GPU workload management
Join the ML Platform team at Synthesia to design infrastructure that supports researchers and product teams. You will build systems for training, evaluating, and deploying generative models, focusing on automation and agent-oriented workflows.
This hands-on individual contributor role requires a systems mindset to improve reliability and resource efficiency. You will shape the platform architecture as the company scales its model workloads and tools.
What you'll do
- Design platform systems for model training, evaluation, and production serving
- Build infrastructure to make ML workloads reliable, scalable, and cost-efficient
- Develop internal tools and workflows operable by humans and agents
- Architect model deployment and serving across research and product environments
- Improve scheduling, monitoring, and debugging of GPU and cloud workloads
- Create abstractions to reduce operational overhead for engineers
- Enhance observability, automation, and developer experience
- Collaborate with researchers to turn pain points into platform capabilities
What you'll need
- Strong experience building or operating reliable, scalable production systems
- Systems mindset focused on bottlenecks, failure modes, and resource usage
- Hands-on experience with cloud infrastructure, Linux, and automation
- Experience with Kubernetes and operating distributed workloads in production
- Strong coding skills, ideally in Python or similar backend languages
- Judgment on balancing automation leverage with human control
- Experience building internal platforms or developer tooling
- Comfort taking ownership of open-ended technical problems
Nice to have
- Experience operating ML infrastructure or model serving systems
- Background supporting research or data-intensive workloads
- Experience with GPU-based systems or performance-sensitive infrastructure
- Familiarity with Terraform, Datadog, or GitHub Actions
- Experience building agentic or LLM-powered internal tools
- Knowledge of workflow orchestration systems like Temporal
- Familiarity with performance optimization or resource allocation
Stack
- Python
- Kubernetes
- Linux
- Terraform
- Datadog
- GitHub Actions
- Temporal
- GPU infrastructure
How you'll work
- Fully remote position
- Teams located across Europe and the US