Avahitech.com is now Avahi.ai
Contact Us

Reinventing Notarial Automation with AI on AWS

Client

Signot

Location

Mexico City, Mexico

Industry

Legal Technology

Services & Tech

Amazon S3, Amazon SageMaker, Amazon Bedrock [Nova Pro, Titan Embeddings], AWS Lambda, Amazon API Gateway, Amazon OpenSearch Service, Amazon DynamoDB, IAM, CloudWatch

Project Overview

Signot, a legal tech innovator in Mexico, partnered with Avahi to explore AI-driven automation for structuring notarial documents. The goal was to build a section classification system that could intelligently identify legal-functional structures within formal Spanish-language documents such as property deeds. Avahi designed and implemented a generative AI-based solution on AWS that successfully demonstrated automated structure detection, human-in-the-loop refinement, and scalable performance. This initiative laid the groundwork for broader transformation of legal workflows.

About the
Customer

Signot is a Mexico-based legal technology company focused on modernizing notarial practices. Its mission is to streamline notarial document preparation and review using AI and automation, enhancing accuracy, efficiency, and profitability for notaries operating under strict legal formalities.

The
Problem

Notarial documents in Mexico must follow strict legal and structural conventions. Manual processing of these records is time-consuming, error-prone, and resource-intensive. Signot needed a system capable of automatically analyzing, labeling, and classifying different sections of legal texts to support downstream use cases, such as clause verification, legal summarization, digital search, and in general text improvement.

Without automation, Signot faced barriers to scaling its services and risked falling short of its vision to modernize the legal workflow. Furthermore, capturing expert feedback and improving system accuracy within a compliant and controlled framework, was critical for productionreadiness.

Why AWS

Signot chose AWS for its scalable, secure, and AI-ready infrastructure. AWS services like Amazon SageMaker and Amazon Bedrock enabled rapid model prototyping, training, and deployment, while serverless components such as AWS Lambda and Amazon API Gateway ensured cost-effective, event-driven processing.

Amazon OpenSearch was also leveraged to support semantic search, storage of classified labels, and document retrieval, all within a compliant and auditable architecture.

Why Signot Chose Avahi

Avahi, an AWS Premier Tier Services Partner, brought deep expertise in generative AI, NLP, and cloud-native design. With a proven ability to build domain-specific AI applications, Avahi collaborated closely with Signot’s legal experts to bridge the gap between technical feasibility and legal accuracy.

Avahi’s phased approach, tailored to Signot’s specific use case, ensured a structured delivery aligned with the legal rigor and timeline expectations of the project. Weekly validations with legal SMEs ensured the classification system met practical notarial needs.

Solution

Avahi delivered a four-week project using a modular AWS architecture that enabled rapid development and refinement of a legal document classification system:

  • Week 1 focused on discovery, document segmentation, and category definition. A dataset of notarial property deeds was ingested, manually labeled, and prepared for model training.
  • Week 2 involved feature engineering using n-grams, document position metadata, and Named Entity Recognition (NER). An initial model was trained with support for weak supervision to expand label coverage.
  • Week 3 introduced semi-supervised learning with label propagation and k-nearest neighbors (k-NN) to improve generalization across unseen documents.
  • Week 4 incorporated legal expert feedback into model refinement. A live demo and documentation concluded the engagement.

Key Deliverables

  • Legal-functional schema and sample labeled dataset
  • A feature extraction pipeline combining statistical and semantic methods
  • Initial and refined classification models
  • Semi-supervised learning integrations
  • End-to-end processing pipeline with validation reports
  • Expert-reviewed and corrected outputs
  • Final demonstration and roadmap recommendations

The end-to-end system leveraged:

  • Amazon S3 for storing documents, artifacts, and outputs
  • AWS Lambda for orchestrating preprocessing and real-time workflows
  • Amazon SageMaker for model training and inference
  • Amazon Bedrock (using Nova Pro and Titan Embeddings) for OCR and paragraph embeddings
  • Amazon OpenSearch for semantic retrieval and clause similarity
  • Amazon DynamoDB for structured label and metadata storage
  • Amazon API Gateway for external triggers and feedback collection

Project
Impact

Signot now has a validated AI system that reliably classifies legal document structures, enabling future automation of document review and clause identification. This will accelerate notarial workflows and reduce manual errors while maintaining legal compliance.

Metrics

  • Successfully classified paragraphs into +25 predefined legal sections and subsections
  • Achieved consistent generalization across unseen legal documents using semi-supervised methods
  • Enabled structured expert feedback for iterative refinement
  • Delivered a scalable, cloud-native pipeline using AWS-native services
Explore Our Services

We highly recommend Avahi as a reliable and innovative technology partner. Their expertise in cutting-edge technologies was instrumental in building our Proof of Concept (PoC) and developing our Minimum Viable Product (MVP). Avahi consistently delivered high-quality solutions on time while maintaining a collaborative, responsive approach. They went beyond expectations by identifying opportunities for enhancement, ensuring scalability and compliance for our law enforcement-focused products. Avahi is the clear choice if you need a tech partner with industry knowledge, professionalism, and a commitment to innovation.

Brandon Puhlman

Founder, Bravo Foxtrot

Ready to Transform Your Business with AI?

Book Your Free Ignition AI Workshop

Let’s explore your high-impact AI opportunities together in a complimentary half-day session
Book Your Workshop

View Our Case Studies

See how we’ve delivered measurable results for businesses like yours
View Case studies