Responsibilities
- Design and build scalable ML infrastructure for deploying and maintaining AI agents in production.
- Automate LLMOps and MLOps workflows, ensuring seamless model training, fine-tuning, deployment, and monitoring.
- Optimize GPU and cloud compute workloads, improving efficiency and reducing latency for large-scale AI systems.
- Develop Kubernetes-based solutions, including custom operators for ML model orchestration.
- Improve system observability and reliability, implementing logging, monitoring, and performance tracking for AI models.
- Work with ML and engineering teams to streamline data pipelines, model serving, and inference optimizations.
- Ensure security, compliance, and reliability in AI infrastructure, maintaining high availability and scalability.
- Participate in on-call rotations, ensuring 24/7 reliability of critical AI systems.
Requirements
- 5+ years of experience in software engineering, MLOps, or infrastructure development.
- Strong expertise in Kubernetes and experience managing containerized ML workloads.
- Deep understanding of cloud platforms (AWS, GCP, Azure) and distributed computing.
- Proficiency in Python, with experience developing services for ML/AI applications.
- Experience with ML model deployment pipelines, including model serving, inference optimization, and monitoring.
Nice to Have
- Familiarity with vector databases, retrieval systems, and RAG architectures is a plus.
- Experience with LLMOps, fine-tuning, and deploying large-scale AI models.
- Experience with GPU workload optimization, ML model parallelization, or distributed training strategies.
- Experience building infrastructure for AI-powered applications.
- Contributed to open-source MLOps tools or AI infrastructure projects.
- Thrive in a fast-moving startup environment and enjoy solving complex technical challenges.
Work Arrangement
Remote (Worldwide) — San Francisco
