Staff Software Engineer, Agentic Search & Retrieval

Datasite • New York City, NY

1d • Hybrid

About The Position

Datasite is hiring a hands-on Staff Software Engineer to own the retrieval layer that supplies context to AI agents working on live deals. The role focuses on ensuring agent answers are grounded in the right, current documents and data while respecting permissions. The engineer will work primarily in Python and partner closely with Product and Data Science, remaining hands-on in technical decisions that shape what agents can access and know.

The position is based in New York City or Minneapolis. Datasite operates as a hybrid company, with in-office collaboration on Mondays, Tuesdays, and Thursdays.

Requirements

• At least 6 years of experience building and operating production software, with a record of leading complex technical work across teams.
• Strong hands-on Python experience, including backend services, APIs, and large-scale data pipelines.
• Experience with production LLM applications that depend on retrieval, such as tool calling, RAG, embeddings, vector search, or context management.
• Strong foundations in permissions, multi-tenant data, and access control for agent retrieval.
• Clear product and technical judgment, including the ability to balance answer quality, latency, and cost and explain tradeoffs to engineering and product leaders.

Nice To Haves

• Experience with long-running agent workflows that search or read documents.
• Experience with a vector store at production scale.
• Experience processing large confidential documents and data tables.
• Experience with hybrid search, citations, and evaluation of retrieval quality.
• Experience with virtual data rooms, M&A documents, or similar document-heavy work.
• Experience with tool interfaces that expose search to an agent.

Responsibilities

• Own retrieval for agents working on live deals, grounding answers in appropriate documents and data while balancing quality, speed, and cost.
• Scale platform processing of large documents and data tables so agents can find and reason over content that cannot fit in one pass.
• Shape the context agents rely on and evolve it as deal workflows become longer and more complex.
• Build search tools for agents that cover both indexed content and content not yet indexed, returning trustworthy snippets, citations, and locations.
• Keep agent context aligned with the data room and users' access permissions so agents cannot read content a user cannot access.
• Determine when a model may answer from retrieved context and when the product must verify the result.
• Mentor engineers and set technical direction for retrieval across teams.

Benefits

• Health insurance, including medical, dental, and vision coverage.
• Retirement savings plan.
• Paid time off.
• Other employee benefits, with specific details provided during the interview process.
• Position may be eligible for bonuses, commissions, or overtime if applicable.

What This Job Offers

• Job Type: Full-time
• Career Level: Senior
• Education Level: No Education Listed