Staff Vector DB Engineer — Build Scalable AI Retrieval

Anthropic Limited; New York (NY); On-site; USD 320,000 - 485,000; Full time

Job summary

Anthropic is seeking a Staff+ Software Engineer to build and scale the vector search infrastructure behind Claude. You will own the vector database layer, shape architecture, and ensure it runs at production scale with real users. You will also influence how recall and latency trade-offs are managed across systems.

You will work in the Databases organization, collaborate with Search Platform and Memory & Search teams, and mentor peers while shaping technical direction for a growing team.

Qualifications

• Experience building vector databases or approximate nearest-neighbor search systems in production.
• Knowledge of ANN index structures (graph-based, clustering-based or disk-resident).
• Experience with search relevance and ranking.
• Strong systems programming skills in Rust, C++, or Go, including performance-critical code.

Responsibilities

• Own the core of Anthropic's vector database layer: index structures, storage layout, query execution, serving layer.
• Ensure production readiness: benchmarking, capacity planning, onboarding new workloads, operational tooling.
• Drive recall, latency and cost trade-offs: quantization, filtered search, index build/refresh strategy.
• Scale throughput to meet rising research/training demands; assess bottlenecks.
• Extend retrieval stack toward hybrid retrieval: lexical search and fused ranking with vector search.
• Set evaluation methodology: recall targets, latency, cost per query.
• Lead design reviews and mentor engineers across teams.
• Share on-call responsibilities for owned systems.

Skills

Vector database

ANN index structures

Search relevance

Systems programming

Education

Bachelor's degree

Job description

Anthropic is seeking a Staff+ Software Engineer to build and scale the vector search infrastructure behind Claude. You will own the vector database layer, shape architecture, and ensure it runs at production scale with real users. You will also influence how recall and latency trade-offs are managed across systems.

You will work in the Databases organization, collaborate with Search Platform and Memory & Search teams, and mentor peers while shaping technical direction for a growing team.