Senior AI Engineer — Legal Document Intelligence

Marco Russo

Full-time · Senior · New_York

This role is no longer accepting applications

Similar open roles you can apply to now:

About the role

We build contract analysis software used by in-house legal teams at 340 companies across financial services, life sciences, and technology. Our platform extracts obligations, rights, termination conditions, and defined terms from commercial agreements and presents them in structured formats that lawyers can review, compare, and export into matter management systems. The AI layer of our product is what distinguishes us from a document management system. We use a combination of fine-tuned extraction models, structured generation via LLMs, and a retrieval layer over the contract corpus. It works well for the document types our training data covers well. It works less well at the edges — unusual clause structures, non-standard defined terms, and multi-party agreements with complex cross-references. The Senior AI Engineer we are hiring will own the technical architecture of this layer and lead the work to extend its coverage and improve its reliability. This is not a prototyping role. The output of this system feeds legal decisions at companies where accuracy matters commercially and reputationally.

Responsibilities

Own the technical architecture of the AI extraction and generation layer — document extraction models, structured generation pipeline, and retrieval infrastructure
Extend extraction coverage to new clause types and contract structures identified as high-priority by the product team and customer success data
Build and maintain evaluation frameworks that measure extraction quality in terms meaningful to legal review: precision on obligation extraction, recall on defined term identification, and consistency across counterpart document versions
Work with legal domain experts to annotate training data and validate model outputs
Contribute to technical due diligence conversations with enterprise customers evaluating the platform

Requirements

5+ years of ML or AI engineering with at least 3 years building production document understanding or information extraction systems
Python at a senior engineering level — production-quality code, clean abstractions, and the discipline to build systems that legal engineers can audit
Hugging Face Transformers for fine-tuning extraction and classification models on legal document datasets
LangChain or LlamaIndex for structured generation and retrieval workflows over contract corpora
RAG pipeline design with specific attention to retrieval precision on highly specialised legal vocabulary and clause structures
SQL for working with contract metadata, extraction results, and evaluation dataset management
Experience with named entity recognition, relation extraction, or span classification in structured document contexts is directly relevant

Benefits

$142,000 – $172,000 base salary + equity (Series B, meaningful strike price)
Full remote — US and UK preferred for time zone overlap with customer accounts
25 days PTO plus company holidays
$2,500 annual learning budget
Direct access to legal domain experts and enterprise customer feedback that will inform your engineering priorities