Senior/Staff Software Engineer, Search & Retrieval Infrastructure

About Pinecone

Pinecone is the knowledge infrastructure for AI at scale. Its leading vector database and knowledge engine, Pinecone Nexus, power accurate, performant AI applications for more than 9,000 customers and 800,000 developers worldwide. Pinecone's mission is to make AI knowledgeable. Pinecone is based in New York and raised $138M in funding from Andreessen Horowitz, ICONIQ, Menlo Ventures, and Wing Venture Capital.

About the Team and Role:

We are hiring a senior/staff software engineer to help design and build core components of our next-generation knowledge retrieval system built for the AI era – search and retrieval infrastructure that powers high-quality, scalable, and enterprise-grade agentic systems. You’ll build the framework that allows our customers to connect knowledge–synthesized from structured and unstructured data–to modern LLM-powered applications, leveraging the world’s best-in-class vector DB supporting semantic search and hybrid retrieval. This role is ideal for someone who loves backend system architecture, distributed systems, and applied AI infrastructure. It is a high impact role with significant ownership across architecture, performance, and system reliability.

Responsibilities:

• Design and build scalable platform components leveraging advanced retrieval via query planning, semantic and hybrid search, metadata-aware search, and LLM generation

• Design and build optimized indexing pipelines for structured and unstructured data

• Build backend services for semantic and hybrid retrieval, knowledge graph construction, and retrieval orchestration

• Improve retrieval quality through evaluation and observability frameworks

• Design APIs for internal and external user and agentic consumers

• Optimize latency, throughput and cost across large-scale inference and retrieval workloads

• Drive technical direction for reliability and security

What You’ll Bring to the Table:

To thrive in this role, you don't need to check every single box, but you should be deeply passionate about how to turn data into knowledge.

Systems Expertise

• Architectural Depth: You have a proven track record (typically 6+ years) of shipping production-grade backends for large-scale systems. You don’t just write code; you design for high throughput, low latency, and long-term maintainability.

• Data Engineering Savvy: You’re comfortable building high-throughput indexing pipelines that handle both the messy world of unstructured data and the rigid world of structured schemas.

AI & Retrieval

• Retrieval Intuition: You understand that "search" is more than just a keyword match. You have direct experience (or deep theoretical knowledge) in semantic search, vector databases, hybrid retrieval strategies, or with traditional search engines like Elastic or OpenSearch.

• RAG & Orchestration: You understand the nuances of Retrieval-Augmented Generation (RAG) patterns, from embedding pipelines and hybrid search techniques to how query planning and metadata filtering can make or break an LLM's performance.

Technical

• Language Fluency: You are an expert in at least one major language like Go, Rust, C++, Java, or Python.

• Infrastructure: Familiarity and experience with modern infrastructure tools, such as Kubernetes, cloud-native architectures, and observability frameworks, as well as infrastructure-as-code tools like Terraform or Pulumi.

Ownership & Impact

• Product Thinking: You don't just build to spec; you build for the user. You can design clean, intuitive APIs that both human developers and autonomous agents will love.

• Ambiguity Navigator: You’re comfortable in a high-growth environment. You prefer "owning a problem" over "executing a ticket."

Bonus Points

• Experience building multi-tenant SaaS platforms.

• Experience with retrieval evaluation frameworks—knowing how to actually measure "good" search results.

• Experience with query planning or agentic reasoning loops (e.g., teaching a system how to break down a complex prompt into multiple specific steps).