Publication — Forthcoming 2026
LLM Essentials.
Michael Bommarito, Daniel Katz
A concise technical reference covering large language model architecture, training, fine-tuning, and deployment — written for technology teams in legal and professional services.
Topics
The Technical Foundation
Everything a legal technology team needs to understand about LLMs — from transformer architecture to production deployment — without the hype.
How transformers work from the ground up: attention mechanisms, positional encoding, layer normalization, and the architectural choices that differentiate modern models.
Pre-training, instruction tuning, and RLHF. What training data matters, how models learn, and why data quality determines model quality — with specific relevance to legal data.
LoRA, QLoRA, and other parameter-efficient approaches. When to fine-tune, when to use RAG, and how to evaluate the trade-offs for legal and professional services applications.
Prompt engineering, chain-of-thought reasoning, tool use, and multi-agent orchestration. The practical techniques for getting reliable output from LLMs in professional contexts.
How to measure LLM performance for legal tasks: benchmarking approaches, human evaluation protocols, automated quality metrics, and the limitations of each.
Inference optimization, serving infrastructure, cost management, monitoring, and the operational considerations for running LLMs in production environments.
Audience
For Legal Technology Teams
This book bridges the gap between academic ML literature and the practical needs of technology teams building LLM-powered applications in legal and professional services. It assumes programming experience but not deep ML background.
Legal technology engineers
Software engineers building LLM-powered features for legal applications who need to understand the technology beneath the API.
Technical architects
Architects evaluating LLM strategies, making build-vs-buy decisions, and designing systems that integrate LLMs into existing legal technology stacks.
Data scientists and ML engineers
Practitioners moving from traditional ML into LLM-based systems who need a concise reference on the specific techniques and trade-offs.
Technical leaders
CTOs, VPs of Engineering, and Directors of Innovation who need enough depth to make informed technology decisions and guide their teams.
Training Programs
Learn These Concepts Hands-On
The material in this book forms the foundation of our LLM Development training track — a 16-hour program designed for senior engineers building with LLMs in legal technology environments.
AI Training ProgramsComing 2026
In Development
"LLM Essentials" is currently in development. Sign up to be notified when it becomes available.
Get NotifiedStay ahead of AI in professional services.
Industry insights, market shifts, and what we're building — delivered monthly.
We won't send you spam. Unsubscribe at any time.