LLM Platform Engineer

About the Role

Melbourne AI is looking for an LLM Platform Engineer to build and operate the AI infrastructure underpinning our client engagements. This role focuses on the platform layer, model serving, vector database management, prompt versioning, observability, and cost controls. You care about reliability and latency, not just model accuracy.

What You'll Do

• Design and maintain LLM serving infrastructure and deployment workflows
• Build and manage vector database infrastructure using tools such as pgvector and Pinecone
• Implement prompt versioning, A/B testing, and evaluation pipelines
• Set up observability and cost monitoring for LLM workloads
• Support agent engineers with platform reliability and scalability requirements