Document Processing · Enterprise Knowledge Search
Draftwise needed to help lawyers draft and negotiate contracts faster by grounding answers and clause suggestions in clients' contracts, communications, and related documents, while still meeting the legal market's expectations for privacy, security, and highly accurate retrieval.
Reported impact
Draftwise improved search-result quality by 30%, saw a threefold increase in API usage in the first quarter of 2025, and gave lawyers more time for strategic client work by reducing the manual review and repetitive drafting that had turned many into 'cut-and-paste drones.'
Draftwise built its Smart Draft contract-drafting product on Cohere models delivered through Microsoft Azure AI Foundry. The workflow uses a retrieval-augmented generation architecture so legal teams can search their own contracts, communications, and prior agreements instead of relying on generic model output. Cohere Embed creates semantic representations across source material, including images, diagrams, tables, and graphs, so lawyers can retrieve precedent by meaning rather than keywords. Cohere Rerank then orders the candidate passages to put the strongest clauses and examples on top, and Cohere Command generates drafting language in a natural interface while citing source documents. Running the stack on Azure let Draftwise combine trusted infrastructure with enterprise-grade search, grounded generation, and multilingual legal support inside one contract workflow.
Client contracts, communications, and related legal documents are indexed semantically with Cohere Embed; when a lawyer asks Smart Draft for a clause or draft, Azure AI Foundry retrieves candidate passages, Cohere Rerank reorders them by relevance, and Cohere Command produces grounded contract language with citations back to the original material.
Cohere boosted search quality by 30%, giving lawyers the right proprietary data at their fingertips.
Draftwise needed a strong retrieval-augmented generation (RAG) system to power its contract generation tool, Smart Draft.
Draftwise selected Cohere's models for their quality, accuracy, and search retrieval capabilities, and subsequently built its RAG system using three powerful Cohere models: Command, Embed, and Rerank.
Embed can process all data, including images, diagrams, tables, and graphs
a threefold increase in API calls made by Draftwise to models over just the first quarter of 2025
Review how this reference was scored and trace each normalized claim back to its documented source.
Scoring methodology
Customer Reference Value measures usefulness as a customer reference. Evidence Confidence measures the quality and traceability of available evidence, not independent verification.
Traceability
Field coverage shows which facts each source supports. Dates are shown only when documented by the source.
The page meta description says: 'Founded in 2020 by James Ding, Emre Ozen, and Ozan Yalti, Draftwise has 50 employees based in New York City.'
Related
MongoDB Sample Sep 9, 2026 Squary AI Squary AI needed to eliminate slow, outage-prone legal search infrastructure that combined MongoDB with OpenSearch and hurt document review speed, cost, and trust. At Quatrini Law Group, Squary AI analyzed more than 460,000 pages and saved over 400 attorney and paralegal hours. Amazon Bedrock · Claude · MongoDB Atlas 
In-house Sample Sep 1, 2026 Thomson Reuters General-purpose LLMs can produce subtle but consequential errors in legal and compliance work, where professionals need trustworthy, citation-backed analysis across very large document sets. As of Q1 2026, CoCounsel has crossed one million users CoCounsel · Practical Law · Tabular Analysis 
OpenAI Sample Aug 20, 2026 Stampli Stampli needed to launch complex new products and keep product knowledge current without waiting on scarce design, contractor, and operations bandwidth. 243 modeled active role-hours compressed to about 77 for the Deep Finance launch ChatGPT Work · Codex · Deep Finance 
