ByteDance
Software Engineer - Search & Vector Database Infrastructure
San Jose, California · Full-time · Posted Oct 28, 2026
- Education
- B.S. in Computer Science, AI, Machine Learning, or a related field.
About the job
Develops and optimizes Elasticsearch/OpenSearch and Milvus products for scalable full-text, vector, and hybrid search, including distributed retrieval and cloud-native search architectures.
What you'll do
- Build high-performance and scalable search and vector retrieval capabilities, including hybrid, filtered, and multi-vector search and large-scale ANN retrieval.
- Optimize vector indexing and distributed retrieval across algorithms, storage, caching, and I/O.
- Explore cloud-native Vector Lakebase architectures for search over data lakes and object storage.
What they ask for
- B.S. in Computer Science, AI, Machine Learning, or a related field.
- Strong software engineering skills in C++, Golang, Java, or Python.
- Experience with search engines, databases, storage systems, or distributed systems; hands-on experience with Elasticsearch/OpenSearch or Milvus in production or system development.
- Strong understanding of distributed systems and performance optimization.
Nice to have
- Deep experience with Elasticsearch/OpenSearch, Lucene, Milvus/Knowhere, or similar systems; knowledge of full-text and vector indexing algorithms such as HNSW, IVF, or DiskANN.
- Experience optimizing billion-scale vector retrieval or with cloud-native storage, compute-storage separation, object storage, or lakehouse/lakebase architectures.
Tools
Elasticsearch, OpenSearch, Milvus, C++, Golang, Java, Python, Lucene
AI focus
The team builds search infrastructure for GenAI applications and Agentic Search products, including RAG, AI-powered search, and multimodal retrieval; this role develops vector and hybrid retrieval capabilities supporting these use cases.
Model-written summary · the original posting is authoritative