Staff AI Infrastructure Architect

What You Will Lead & Build

• Design zero-leakage enterprise knowledge meshes indexing multi-terabyte vector datasets.
• Implement hybrid sparse-dense retrieval pipelines with sub-50ms latency SLAs.
• Optimize local LLM inference engines (vLLM, TensorRT-LLM) on private GPU hardware.
• Establish automated evaluation frameworks for hallucination prevention and citation accuracy.

What We Look For

• 7+ years of software engineering with 3+ years dedicated to enterprise AI / ML infrastructure.
• Deep expertise with vector databases (Qdrant, Milvus, Pinecone) and embedding fine-tuning.
• Solid understanding of cryptographic privacy, role-based access tokens, and data residency.
• Strong Python and C++/Rust systems background.

Core Technologies & Tools