About the role
Synthesia is hiring a Principal ML Platform Engineer to build and operate the systems that let research and product teams train, evaluate, serve, and deploy generative models. The role spans research infrastructure, production serving, internal tooling, and platform interfaces, with a focus on reliability, scalability, performance, and efficient use of resources. It is a hands-on individual contributor position with significant ownership of how the platform evolves as model workloads and teams grow.
You’ll shape platform architecture, improve GPU workload scheduling and observability, and create automation-friendly tools and workflows for researchers and engineers. Synthesia is looking for strong production systems experience, cloud infrastructure, Linux, Kubernetes, and coding skills such as Python. Experience operating ML infrastructure, model serving, or data-intensive and GPU-based workloads is particularly relevant. The role is remote in Europe, with the UK also listed as an ATS.
Highlights
- Design platform systems for model training, evaluation, and production serving.
- Improve reliability, scalability, cost efficiency, monitoring, and debugging.
- Build internal tools and workflows for engineers, researchers, and agents.
- Work across cloud infrastructure, Kubernetes, distributed systems, and GPUs.
This recap is dataskew's editorial summary, not the company's copy.