5 hours ago
Base Salary
$226k - $283k/yr
Responsibilities
- Build developer tooling and MLOps capabilities on Ray for developers and coding agents.
- Develop CLI, SDK, API, and MCP surfaces with self-discovery, structured errors, dry-run support, and consistent platform behavior.
- Improve Anyscale Workspaces workflows covering environments, dependencies, images, authentication, workload submission, and debugging.
- Build tools for data preparation, fine-tuning and post-training, evaluation, production serving, dataset management, experiment tracking, and lineage.
- Develop model registration, deployment, benchmarking, and LLM service-metrics capabilities for production endpoints.
- Provide observability and failure diagnosis across jobs, tasks, actors, nodes, and GPUs.
- Design integrations using OpenAI-compatible APIs, OpenTelemetry, and stable Jobs and Services interfaces.
- Design and operate highly available backend services and platform architecture across serverless and bring-your-own-cloud environments.
- Work with users, field teams, distributed systems experts, and machine learning experts to scope, ship, and iterate on products.
Requirements
- Bachelor's degree in Computer Science, Engineering, or equivalent practical experience.
- At least 5 years of experience writing high-quality production code.
- Strong foundation in algorithms, data structures, and system design.
- Experience with modern machine learning tooling, including PyTorch, MLflow, data catalogs, or similar tools.
- Hands-on experience building and operating highly available production services.
- Strong product instincts and a record of shipping developer-facing tools that users choose to use.
- Preferred: experience building and maintaining open-source projects.
- Preferred: experience operating machine learning infrastructure in production.
- Preferred: experience building highly available serving systems.
- Preferred: experience using Ray.
Benefits
- The posting identifies Anyscale as an Equal Opportunity Employer.
- Anyscale is an E-Verify company.
Tech Stack
MLflowPyTorch
Categories
About Anyscale
Anyscale enables Python developers to build and run all their AI—from data prep to training and inference—at any scale. Anyscale is trusted by leading AI teams at Canva, TripAdvisor, Physical Intelligence, Coinbase and more.