AI for legal : Developing a legal AI ecosystem
Details
**** Law is one of the highest-stakes environments for AI. A “hallucination” here isn’t just okay—it’s a malpractice lawsuit. In this session, we move beyond the hype of AI and tools to discuss the engineering required to build a Legal AI Ecosystem.
We will cover how to architect systems that respect “Ethical Walls,” why RAG (Retrieval-Augmented Generation) is the bare minimum, and how to select a tech stack that handles strict data residency requirements.
What We’ll Cover:
- Design First: The “Ethical Wall” architecture and data privacy by design.
- The Tech Stack: Vector DB and RAG, LLMs (Claude vs. GPT vs. open source), and Orchestration (LangChain vs. specific agents).
- The Important Stuff: Solving for hallucinations, citation accuracy, and handling millions of documents.
- Pitfalls: Why generic AI wrappers are a great start in LegalTech and how do you add more value
About Speaker:
- Ashwin Phadke - Sr. Manager, AI and Machine Learning @ Servient | Generative AI | LLM | Computer Vision | Deep Learning | Autonomous Vehicles | Prev - Amazon
