remote
Senior Software Engineer, Vector Index Research - zilliz
Software Engineer
Lead the design and implementation of high‑performance vector retrieval engines, optimizing similarity search over massive embedding datasets using Python, C++, and Rust in a distributed, GPU‑accelerated environment.
About the role
Key Responsibilities
- Architect and develop core vector index components for Milvus, Zilliz Cloud, and Vector Lakebase, ensuring sub‑millisecond query latency at scale.
- Optimize similarity search algorithms (e.g., IVF, HNSW) for CPU and GPU execution, leveraging SIMD and CUDA where appropriate.
- Collaborate with data scientists to integrate new embedding models and evaluate retrieval quality metrics.
- Design fault‑tolerant, sharded index structures that support horizontal scaling across clusters.
- Conduct performance profiling, benchmark testing, and continuous refactoring to maintain top‑tier throughput.
Requirements
- 5+ years of systems programming experience in Python, C++, or Rust.
- Deep knowledge of vector search algorithms, ANN indexing, and similarity metrics.
- Hands‑on experience with distributed systems, message queues, and container orchestration (Kubernetes).
- Proficiency in GPU programming (CUDA/OpenCL) and SIMD optimizations.
- Strong analytical skills, ability to translate research papers into production‑ready code.