Lead AI Engineer / AI Solutions Architect – Build a Private AI-Native Legal Operating System

Description Our litigation law firm is looking for an experienced AI engineer or AI architect to build a private, secure AI platform that serves as the firm's internal operating system. This is not a chatbot project. The goal is to build a sophisticated AI platform that understands every case in our firm, assists with legal drafting, manages documents, performs legal research, automates client communications, and integrates with our existing software. We are looking for someone who has already built production AI applications—not someone who is experimenting with LLM APIs. This is expected to become a long-term project. About Our Firm We are a busy litigation firm handling: Wage & Hour Litigation Personal Injury Wrongful Death Lead Paint ADA Litigation Employment Litigation We generate thousands of pages of pleadings, discovery, depositions, medical records, expert reports, and correspondence. The platform must become the firm's "second brain." Vision Every case should have its own AI workspace. The AI should understand every document in that matter and be able to answer questions such as: Summarize the liability evidence. Draft a motion to compel. Identify contradictions between deposition testimony and interrogatories. Find missing discovery. Draft a client update. Build a timeline. Locate every document discussing a specific witness. Draft deposition outlines. Prepare trial examinations. The system should continuously improve as additional documents are uploaded. Core Features Matter Management One AI workspace per case Matter timeline Parties Witnesses Experts Deadlines Tasks AI Document Intelligence OCR Automatic indexing Semantic search Timeline generation Issue spotting Document summaries Duplicate detection Exhibit organization Drafting The AI should draft: Complaints Answers Discovery demands Responses Motions Memoranda of law Letters Client communications Emails Deposition outlines Trial outlines Claude should be the primary drafting model. Research The platform should: Search internal firm documents Search uploaded pleadings Search legal authority (future phase) Build legal research memoranda Calendar Integration with: Google Calendar The AI should understand: Discovery deadlines Court conferences Hearings Depositions Statutes of limitation Client Communication Integration with Gmail. Eventually: Draft emails Summarize client communications Translate English ↔ Spanish Track communications Suggest follow-ups Integrations Current systems include: Google Workspace Gmail Google Calendar Dropbox SmartAdvocate GoHighLevel CallRail QuickBooks (future) Security This is confidential legal data. The application must include: End-to-end encryption Role-based permissions Audit logging Secure authentication Encrypted document storage No client data used for model training Preferred Technology Stack We are open to recommendations, but currently envision: Frontend React Next.js TypeScript Backend Python FastAPI Database PostgreSQL Vector Database pgvector (or equivalent) Storage AWS S3 AI Anthropic Claude OpenAI (secondary) RAG Embeddings Agentic workflows Infrastructure Docker AWS Experience Required Please apply only if you have experience building production AI systems. Ideal experience includes: Claude API OpenAI API RAG architectures Vector databases LangGraph or similar orchestration frameworks Semantic search Document ingestion pipelines OCR AI agents Enterprise security SaaS applications Legal technology experience is a major plus. What We Need From You Please include: Similar AI systems you have built. Your recommended architecture. What you would change about our proposed design. How you would structure long-term memory across thousands of documents. Estimated timeline for an MVP. Estimated cost. Your availability. Project Scope This is intended to become the central operating system for our law firm. The successful applicant will likely become our long-term AI technical lead as additional features are added over time.