Question

A legal services company wants to build an internal application where staff upload contract PDFs and receive concise summaries plus answers to natural-language questions about each document. The team has no ML engineers and wants to minimize custom model development and infrastructure management while grounding responses in the uploaded documents. Which approach best meets these requirements?

Accepted Answer

Use the Gemini Enterprise Agent Platform with a grounded generative agent connected to the uploaded documents as a data store, exposing summarization and Q&A through the platform's managed APIs.

Answer

Train a custom transformer model on Vertex AI using the company's historical contracts, then deploy it to a GPU-backed endpoint for inference.

Answer

Store the PDFs in Cloud Storage and run a nightly Dataproc Spark job that extracts text and applies open-source summarization libraries.

Answer

Use the Cloud Vision API to extract text from the PDFs and store the raw extracted text in BigQuery for staff to query with SQL.