AI Prompt Engineering Framework

8 posts · Jul 2025 – Jun 2026

What this trail covers

Harvey's platform continues to evolve its AI agent capabilities, shifting the paradigm from prompt-based assistance to task-oriented execution. The system now empowers agents to autonomously plan, research, execute, and deliver finished work products, with lawyers focusing on goal setting and final review. This evolution spans transactional work, litigation, compliance, and in-house operations, aiming to increase leverage by automating repeatable, high-volume tasks. The platform now integrates m

All Harvey trails

Jun 2026How to Use AI to Write Legal Memos More EffectivelyThis post details a five-step workflow for lawyers to use AI more effectively in drafting legal memos. It emphasizes treating AI as a junior drafter, not a research source, and outlines a prompt structure including role, context, source material, and output specification. The workflow highlights the lawyer's role in assembling inputs, drafting prompts, reviewing AI-generated drafts, and performing final verification. It also discusses the limitations of general-purpose AI in legal memo work, particularly regarding citation grounding and matter-level isolation.
Jun 2026How Legal Drafting AI is Changing the Way Lawyers WorkThis post details how AI for legal drafting, specifically grounded legal drafting AI, is changing lawyer workflows. It explains the three layers of such AI: the underlying language model, the grounding layer (RAG), and the workflow surface. It highlights that drafting is a safe entry point for legal AI due to the human review step. The post emphasizes the importance of grounding AI outputs in vetted legal sources and the verification loop, which includes surfacing citations and reasoning. It outlines AI's role across the legal drafting lifecycle: from drafting from a blank page using organizational precedent, to building and pressure-testing briefs with AI-proposed authority, to summarizing dense files and localizing/editing documents. The evaluation criteria for AI legal drafting tools focus on grounding sources, jurisdictional awareness, citation handling, integration, confidentiality, and auditability. The post implicitly reinforces the value of Harvey's approach by detailing these aspects.
Feb 20265 Workflows for IP and Patent LitigationThis post introduces five specific workflows for IP and patent litigation that leverage the AI Prompt Engineering Framework. These workflows demonstrate how to automate tasks such as drafting infringement claim charts, analyzing patent office actions, generating invalidity contentions from prior art, drafting patent license agreements, and preparing filing documents. Each workflow is broken down into specific prompt-based steps, illustrating the practical application of structured prompting for complex legal tasks.
Dec 2025How our In-House Legal Team Uses Harvey to Work SmarterThis post details the introduction and impact of two new features: the 'Improve' button, which uses AI to enhance user prompts, and the 'nudge' feature, which suggests relevant knowledge sources (e.g., Web Search, LexisNexis) to ground AI responses. It provides examples of how these features work and their benefits for prompt quality and user learning.
Jul 2025Integrating Deep Research into HarveyThis post details the rapid integration of OpenAI's Deep Research API into Harvey, achieved in under 12 hours. It highlights the role of Harvey's AI-native architecture, specifically the Workflow Engine, in enabling this speed. The Workflow Engine's composable AI building blocks and 'thinking states' facilitated the seamless integration of Deep Research's reasoning and actions, allowing for user intervention and transparent evaluation of intermediate results. The post also emphasizes the successful incorporation of Deep Research's URL citations into Harvey's existing citation system for traceability and verification. Furthermore, it describes the use of AI-assisted development, including AI coding assistants and Streamlit, to overcome challenges in integrating a new, complex API with dynamic output types and product frameworks.
Advertisement