Thank you for sending your enquiry! One of our team members will contact you shortly.
Thank you for sending your booking! One of our team members will contact you shortly.
Course Outline
Day 1: Establishing the Foundation—Ingest, Search, Retrieve
Module 1: The Legal Engineer’s Landscape
- Learning objectives—understand the role, where AI integrates into legal work, and the two overarching risks.
- Topics
-
- The legal engineer role and the current demand for these skills.
- Where AI fits: eDiscovery, review, contracts, research, and investigations; the EDRM model explained in plain terms.
- Deciding between building custom solutions versus buying existing ones.
- The two critical risks permeating all activities: confidentiality/privilege and defensibility.
Module 2: Handling Messy Legal Data—Ingestion and Extraction
- Learning objectives—manage the realities of legal data at scale.
- Topics
- Managing over 1,400 file types, email and PST formats, scanned paper, and load files (.dat/.opt); identifying embedded metadata that matters.
- Text extraction (Tika), OCR, and de-duplication strategies.
- Lab: FreeEed Ingestion—construct an ingestion pipeline handling a deliberately messy document set (email/PST, scans, load files).
Module 3: Search and Retrieval—The Foundation
- Learning objectives—build the core eDiscovery primitive: finding anything within everything.
- Topics—full-text search and indexing (Solr/Lucene); relevance, metadata, and date filtering; searching across OCR’d content.
- Lab: eDiscovery Search—index a corpus and execute real eDiscovery-style searches, including within OCR’d scans.
Module 4: RAG for Legal Documents—With Citations
- Learning objectives—develop RAG over legal documents that provides source citations.
- Topics
- Why retrieval, rather than fine-tuning, is preferred for sensitive material—the model does not ingest the documents.
- Chunking, embeddings, and crucially citations/provenance.
- Summarization across multiple documents and threads.
- Lab: Legal RAG with Citations—build a RAG Q&A system over a document set that answers queries with source citations.
Day 2: Ensuring Privacy, Defensibility, and Deployability
Module 5: Privacy, Privilege, and Local Serving—The Privilege Trap
- Learning objectives—keep legal data local and ensure it is certifiable.
- Topics
- Tracking where data goes when processed by cloud AI.
- Privilege waiver, duty of competence, and the “private” spectrum (contractual vs. physical isolation).
- Morgan v. V2X and why local processing is court-defensible.
- Serving local models (Ollama / vLLM) and monitoring outbound traffic.
- Lab: Local Model + Egress Proof—run a local model end-to-end and demonstrate, via monitoring, that no data egressed.
Module 6: Defensible AI Review
- Learning objectives—measure and document AI reviews to ensure they hold up in court.
- Topics
- Critical metrics for court: recall, elusion, precision, ground-truth validation; TAR / active learning.
- Transparency (why was this document coded?) and reproducibility—pinning the model, fixing settings, and logging everything.
- Creating a “defensible case snapshot” allowing a review to be re-run a year later with identical results.
- Lab: Defensible Review—measure an AI review against a blind ground truth and generate a reproducibility bundle.
Module 7: Deployment—Workflow, Private Hosting, and Governance
- Learning objectives—assemble components into a workflow, deploy privately, and evaluate performance.
- Topics
- Constructing a multi-step legal workflow (ingest → search → summarize → review → produce) with human-in-the-loop controls.
- Essentials for private/on-prem deployment (containerization; keeping data within the building).
- Overview of AI governance for legal contexts and system scoring using SAIS-100 (the Elephant Scale Secure AI Score).
- Lab: Score and Package—integrate a multi-step workflow, score it with SAIS-100, and package it for private deployment.
Capstone (integrated across Day 2)
- Build a private, defensible legal-AI application end to end—ingest a messy corpus, search it, answer questions with citations using a local model, measure a defensible review, and package it for private deployment.
- Participants will leave with a portfolio project that mirrors the responsibilities of a legal engineer role.
Optional Day 3 / Advanced Modules (delivered as a third day or a modular series)
- Investigations: Entities, Relationships, and Timelines—extract people/orgs/dates, reconstruct email threads, build chronologies, and map near-duplicates and document lineage. Lab: create a timeline and entity/relationship view.
- Agentic and Multi-Step Legal Workflows (Advanced)—richer orchestration, contract analysis, multi-document synthesis, and tool use with guardrails as a design principle. Lab: build a multi-step workflow with human checkpoints.
- Deployment at Scale—on-premises and appliance deployment, distributed processing for high volumes, regulated environments (CJIS, government, higher education), and hardware sizing. Lab: containerize and scale a processing job across workers.
- Governance and Compliance Deep-Dive—the AI regulation landscape (100+ US state AI laws, EU AI Act), audit requirements, and a comprehensive SAIS-100 governance audit. Lab: audit a legal-AI system against a governance/defensibility checklist.
Requirements
- Proficiency with Python and basic APIs.
- Beneficial: Familiarity with LLMs at a user level (no machine learning background is required—we will build the necessary mental models).
- No legal background is required—all necessary legal concepts are taught in context.
Audience
- Software and AI engineers transitioning into legal tech.
- Engineers at legal-tech companies who require deeper legal domain expertise.
- Technically minded professionals in legal, eDiscovery, or information governance who wish to build solutions rather than merely purchase them.
- Any professional aiming for a “legal engineer” or “AI legal engineer” role.
14 Hours
Testimonials (1)
That i gained a knowledge regarding streamlit library from python and for sure i'll try to use it to improve applications in my team which are made in R shiny
