Staff+ Software Engineer, VectorDB

Details

Anthropic is looking for a Staff+ Software Engineer to work on the vector search infrastructure behind Claude. Retrieval supports web search across claude.ai, the API and agentic surfaces, memory across conversations, and reinforcement learning workloads through fast, cost-efficient, high-recall nearest-neighbor search over billions of vectors.

The role is part of the Databases organization and works closely with the Search Platform team, which owns web search indexing, retrieval, and serving, and the Memory & Search team, which owns search and memory inside claude.ai. The engineer will help shape the architecture, performance, scalability, and cost of Anthropic's vector database layer.

Responsibilities

• Own the core of Anthropic's vector database layer, including index structures, storage layout, query execution, and the serving layer.
• Own production readiness through benchmarking against exact search, capacity planning, onboarding new workloads, and operational tooling.
• Drive recall, latency, and cost trade-offs involving quantization, filtered search, index build and refresh strategies, and separation and scaling of compute and storage.
• Scale throughput by an order of magnitude as research and training demand grows, determining whether bottlenecks result from architecture, hardware, or engineering bandwidth.
• Extend the retrieval stack toward hybrid retrieval, including lexical search and fused ranking alongside vector search.
• Establish evaluation methodology, including recall targets, latency objectives, cost per query, and regression detection.
• Set technical direction for a growing team, lead design reviews, make architectural decisions, and mentor engineers.
• Share on-call responsibilities for owned systems.

Requirements

• Direct experience building a vector database or approximate nearest-neighbor search system that ran in production at scale.
• Deep knowledge of ANN index structures, including graph-based, clustering-based, or disk-resident approaches, as well as vector quantization and filtered search.
• Experience with search relevance and ranking.
• Strong systems programming skills in Rust, C++, Go, or a similar language, including performance-critical and memory-intensive code.
• Experience designing and operating large-scale distributed storage or serving systems with strict latency and reliability requirements.
• A track record of independently scoping and delivering ambiguous, multi-quarter technical projects.
• Strong written and verbal communication skills, including experience driving alignment across teams with different priorities.
• Interest in AI safety and Anthropic's mission.

Preferred Qualifications

• Experience with object-storage-backed or disaggregated compute and storage database designs.
• Experience building or operating full-text search engines, including inverted indexes and BM25, or hybrid retrieval systems.
• Familiarity with embedding models and the retrieval demands of LLM training and inference.
• Experience using GPUs or other accelerators for index building or search.
• Prior experience as a technical lead on a database or search infrastructure team.

Education and Experience

• Bachelor's degree or an equivalent combination of education, training, and/or experience.
• A field relevant to the role, demonstrated through coursework, training, or professional experience.
• Years of experience required will correlate with the internal job level requirements for the position.

Benefits

Anthropic offers competitive compensation and benefits, optional equity donation matching, generous vacation and parental leave, flexible working hours, and office spaces for collaboration. Anthropic sponsors visas where possible and retains an immigration lawyer to assist with visa applications.

The role follows a location-based hybrid policy. Staff are currently expected to be in one of Anthropic's offices at least 25% of the time, although some roles may require more office time.