This repo provides Generative AI and AI/ML code samples, blueprints (end-to-end solutions) and proof of concepts oriented to the LATAM market
This repo is organized as follows:
End-to-end solutions for a specific use case.
- They include CloudFormation templates to provision the solution to your account.
- They may include a demo UI.
Here is a list of available blueprints:
| Use Case | Industry | Description | Type | Languages |
|---|---|---|---|---|
| Multipage Document Analysis | Financial Services/Legal/Cross-industry | Use this blueprint to extract key information from lengthy documents in highly regulated industries leveraging Amazon Bedrock's wide array of LLMs. | Backend (Python) + UI (React) | Spanish, English |
| Contract Compliance Analysis | Legal / Cross-industry | This project automates the analysis of contracts by splitting them into clauses, determining clause types, evaluating compliance against a customer's legal guidelines, and assessing overall contract risk based on the number of compliant clauses. This is achieved through a workflow that leverages Large Language Models via Amazon Bedrock. | Python for Backend, TypeScript (React) for Frontend | English (can be customized to other languages) |
| Smart Product Onboarding | Retail CPG | Reimagine the seller product onboarding experience with generative AI to create product titles and descriptions, categorize, and extract attributes. | Backend (Pyhton) + UI (React) | All languages supported by Amazon Nova and Anthropic Claude models |
| RFP Answers With Generative AI | Cross-industry | Leverage existing company documents to automate responses to a Request for Proposal (RFP) document. | Backend (Python) + UI (React) | English (can be customized to other languages) |
| Next generation marketing campaigns with Amazon Nova | Retail CPG/Cross-industry/Marketing | Quickly iterate over ideas for visuals for marketing campaigns and accelerate the creation of visual assets for any marketing campaign with the Amazon Nova family of models. | Backend (Pyhton) + UI (React) | English (can be customized to other languages) |
Code samples/notebooks that demonstrate a specific AI/ML functionality in AWS.
Here is a list of available samples:
| Use Case | Industry | Description | Type | Languages |
|---|---|---|---|---|
| Multi-Agent Collaboration System | Cross Industry | Multi-agent customer support system with escalation capability using AWS CDK and Amazon Bedrock | CDK Python | Multilanguage |
| SQL Agent with Amazon Bedrock | Cross Industry, Business Insights | Build and deploy a SQL Agent that lets users query databases using natural language instead of writing SQL code. The implementation combines enterprise features from Amazon Bedrock with LangChain's specialized SQL tools. | CDK Python | Multilanguage |
| Whatsapp with Agents for Amazon Bedrock | Cross Industry, Customer Experience | Implement a WhatsApp communication channel using AWS End User Messaging for social integration. Integrate your business logic with Amazon Bedrock Agents for AI-powered interactions. | CDK Python | Multilanguage |
| WhatsApp Integration with Amazon Connect (with voice notes) | Cross Industry, Customer Experience | AWS-powered WhatsApp integration with Amazon Connect enabling seamless customer service chat with voice note transcription capabilities. | CDK Python | Multilanguage |
| Amazon Connect Custom Bot with Bedrock Agent and Human Escalation | Cross Industry, Customer Experience | A custom solution that integrates Amazon Connect with Bedrock AI agents, enabling advanced chatbot interactions and intelligent escalation to human agents, optimizing customer service through sophisticated conversation flows or multi-agent reasoning. | CDK Python | Multilanguage |
| Amazon Nova Samples | Cross Industry, Customer Experience | Amazon Nova family of models samples that cover various use cases (text generation, document processing, content generation, agentic workflows, etc). | Jupyter Notebooks | Multilanguage |
| WardrobeGenAI Demo | Retail, Customer Experience | AI-powered fashion recommendation system using AWS Bedrock and Anthropic Claude for personalized style suggestions. Features real-time image analysis and interactive shopping experience. | React + Node.js | TypeScript |
| Claude 3.7 thinking in multimodal | Cross Industry | This project demonstrates Claude 3.7's real-time reasoning processes with text, images, and documents using AWS Bedrock converse API. The repository includes three Jupyter notebooks and a custom interface for observing how the AI model approaches different tasks. | Python | Multilanguage |
| Amazon Nova Sonic Conversation | A demonstration project showcasing Amazon Bedrock's Nova Sonic model for real-time speech processing and natural voice interactions using bidirectional streaming capabilities. | Python | English |
- Code samples/notebooks that aim to prove a concept in Generative AI or AIML.
- Research paper techniques implementations
- Novel tecniques implementations
- May include Cloudformation / CDK
Please refer to the CONTRIBUTING document for further details on contributing to this repository.