Define the long-term technical roadmap for our AI platform, covering everything from model serving and feature stores to experiment tracking and CI/CD for ML.
Establish engineering benchmarks for deployment, versioning, A/B testing, and automated rollbacks.
Lead strategies for GPU/compute efficiency and cost optimization while managing the complexities of LLM inference (quantization, batching, and latency) at scale.
Design sophisticated monitoring and alerting systems specifically tailored for AI workloads in production.
Drive platform stability and partner with application teams to ensure our infrastructure meets evolving product needs.
Mentor Senior engineers, lead architecture reviews, and evaluate the next generation of cloud services and ML frameworks.

Requirements

12+ years of engineering experience, with at least 4 years at a Principal or Distinguished level.
Expert-level knowledge of model serving (e.g., Triton, vLLM, Ray Serve) and deep experience with Kubernetes and cloud ecosystems (AWS/GCP/Azure).
Proven experience with tools like MLflow, Weights & Biases, or Kubeflow.
High proficiency in Python and a "systems-thinking" approach to Docker, Helm, and containerization.
Hands-on experience operating LLM inference at scale and a deep understanding of the trade-offs between throughput and latency.
A track record of building internal platforms that treat other engineers as the primary customer, drastically improving engineering velocity.

Benefits

A competitive base pay
Medical Benefits
Discretionary incentive plan based on individual and company performance
Focus on development: Access to a learning & development platform and opportunity for you to own your career pathways
Flexible work policy

Applicant Tracking System Keywords

Tip: use these terms in your resume and cover letter to boost ATS matches.

Hard Skills & Tools

model servingKubernetescloud ecosystemsPythonDockerHelmLLM inferenceA/B testingCI/CDGPU efficiency

Soft Skills

mentoringleadershiparchitecture reviewsstrategic planningcollaboration