RGDSolutions
← Back to selected work

Morgan & Morgan · AI product engineering

AI document intelligence for legal teams

Ricardo’s work at Morgan & Morgan spanned retrieval, evaluation, and AI-assisted drafting. The application connected document intelligence to practical legal workflows, including demand-letter generation and drafting inside Microsoft Word.

Ricardo’s engineering experience

From documents to useful answers

Ricardo built retrieval-augmented generation with per-answer provenance. Answers were connected to their source material, giving legal staff a way to review the underlying information rather than relying on generated text alone.

Make quality part of the system

An evaluation harness with five judge models supported answer-quality assessment. The work also included agentic tool calling, extending the application beyond retrieving information to performing tasks.

Meet people where they work

Demand-letter generation ran on AWS, while AI drafting was integrated into Microsoft Word. These were distinct parts of the same product engineering challenge: connect AI capabilities to the places where legal work happens.

From information to action

Your documents

Knowledge your team already has.

Retrieve. Check. Respond.

Connect relevant sources, evaluate answer quality, and put the result to work.

Answers with sourcesDrafts to reviewActions to take
Illustrative AI workflow · Not a client product screenshot

Technology: Python · TypeScript · AWS · Elasticsearch · Terraform

Building something similar?

Tell us where you are today and what you need to ship next.

Discuss your project