Data Engineer, AI Backend

Toppan Merrill is hiring a Data Engineer, AI Backend to build and operate the data services, APIs, and pipelines behind Quinn’s AI capabilities for legal and financial disclosure workflows. The role includes SEC filing ingestion and indexing, document-context and citation services, AI model routing, and evaluation data pipelines.

Responsibilities:

• Publish and own the AI backend’s API specification, starting with a working stub, then build and operate the production services behind it, keeping the contract stable and communicating changes clearly to the teams that depend on it
• Build a document-context service that gives the AI the relevant context from the document an attorney is working on, while respecting that attorney’s access permissions
• Build ingestion, processing, and indexing pipelines for SEC filings from EDGAR to support peer-company analysis
• Create a citation index that connects AI-generated answers to the exact source sentences in the underlying filings
• Build services that route requests across multiple AI models while providing logging, monitoring, reliability controls, and visibility into usage and cost
• Develop and maintain the pipelines that produce evaluation data used by data scientists to validate AI quality
• Establish data-quality, observability, and operational standards for the AI backend

Requirements:

• 5+ years of professional data engineering or backend engineering experience in Python
• Experience building and operating production data pipelines at scale using Databricks, Apache Spark, or a comparable platform
• Experience building and operating production Python services (FastAPI or similar), including APIs other engineering teams depend on: publishing, versioning, and maintaining a stable contract
• Experience working with text or document data, search, indexing, or information-retrieval systems
• Strong data modeling, pipeline reliability, testing, and observability practices
• A track record of delivering against firm milestones with cross-team dependencies
• Experience with retrieval-augmented generation (RAG) or vector/hybrid search in production
• Experience building citation, provenance, or source-traceability capabilities for search or AI-generated answers
• Experience with SEC filings, EDGAR, XBRL, or other complex financial and regulatory documents
• Experience with Azure data and AI services, including multi-model routing, usage monitoring, and cost controls
• Experience building evaluation datasets or pipelines for generative AI