Land Registry and Ownership document parser using Artificial Intelligence (AI)

[GENERAL] Name and surname of main applicant

Nilay Saha

[GENERAL] Are you delivering this project as an individual or as an entity (whether formally incorporated or not)

Entity (Incorporated)

[GENERAL] Co-proposers and additional applicants

• Vinh Tuan
• Tra Giang

[GENERAL] Please specify how many months you expect your project to last (from 2-12 months)

12

[GENERAL] Please indicate if your proposal has been auto-translated into English from another language

No

[GENERAL] Summarize your solution to the problem (200-character limit including spaces)

We would like to solve this by using generative AI for parsing the documents, to assist the human verifier. This would include different text formats and languages. The AI will be acting as a copilot.

[GENERAL] Does your project have any dependencies on other organizations, technical or otherwise?

No

[GENERAL] If YES, please describe what the dependency is and why you believe it is essential for your project’s delivery. If NO, please write “No dependencies.”

No dependencies

[GENERAL] Will your project’s output/s be fully open source?

Yes

[GENERAL] Please provide here more information on the open source status of your project outputs

The learning and the actual code output to generate the AI output will be open sourced for the community. 

[METADATA] SDG rating

• Reduced inequality : Now every individual can issue their own verification data via this platform that will ensure property rights globally.
• Sustainable cities and communities : Now communities can use this platform to issue themselves NFT based on their asset requirements.
• Decent work and economic growth: The marketplace will foster job creation in communities globally related to real world asset verification.

[SOLUTION] Please describe your proposed solution

Document AI

As part of the review process on the Reitcircles platform, different players in the ecosystem (legal, owner, notary, government etc.) would upload documents related to the verification process. And each of these documents needs to be reviewed.

Verified (KYC +Qualifications) Human agents are part of the review process. And the AI agent being created will help in parsing and extracting relevant information making the review easier for the agents.

• Following diagram illustrates the overall process of AI based flow of data.
• 

• Also there will be UI/UX which will integrate with the above data pipeline and the backend API that will connect to it.
• So all the above components help in the process of verification on the platform.
• This platform also enables digitization of land registry and make maximum use of AI for this.

With AI integrated into the data being generated on the Cardano network, we will be able to better leverage the Cardano blockchain for real world applications.

[IMPACT] Please define the positive impact your project will have on the wider Cardano community

• The data and the model generated from the Document parsing will eventually feed into the blockchain, as utility will increase from real world projects, 
• The automation of document parsing leads to a faster and more robust verification process and SDKs created for this will benefit other Realfi projects who are dealing with similar challenges.

The impact of AI can be measured in terms of the time and effort the verifiers spend in arriving at the final result. And how over time, the system should self learn and process documents dynamically at the time of upload itself, helping the verifiers with cleaner data sources during review.

The learnings, data models generated, and the AI code, will be made open sourced so that other real world projects in the ecosystem can add to the knowledge base and help expand the utility of the chain.

[CAPABILITY & FEASIBILITY] What is your capability to deliver your project with high levels of trust and accountability? How do you intend to validate if your approach is feasible?

Reitcircles has achieved the following milestones already:

• A working portal with wallet based login
• An onboarding module with document upload and review
• Payment module for taking NFT payments and minting the same.
• House NFT creation based on documents and smart contract based verification.
• Portal based chat communication between members of the platform.
• Role NFT based access to the platform.
• Smart contract based REIT token issuance replicating BTC (with halving every 4 years)
• 
  

Our team brings in more than 40+ yrs of software development experience.

• Antonio Hernandez-Garduno: Platform lead in Smart contract development who is a PhD graduate from Caltech, leading security solutions.
• Tony Mönicke: Computer science graduate based in Berlin and working in cardano ecosystem for last 2+ years.
• Nilay Saha: Worked in leading multinational companies such as ASML, AUDI, Hilti and extensive experience in banking. Has 25+ years of development, solutions architecture and technical leadership experience in large/mid size industry.
• An Sy, Vinh tuan: Developers working on the platform (frontend and backend)
• Giang Nguyen: UI/UX lead

We will be delivering using open source code, where commits can be monitored. And to give a proof of concept / working version, we will integrate the same into our verification platform that is already live on mainnet.

[PROJECT MILESTONES] What are the key milestones you need to achieve in order to complete your project successfully?

Data Source Identification and Collection Setup

• Identify and finalize data sources for land registry information.
• Set up mechanisms to collect data from identified sources.
• Ensure data collection processes are automated and reliable.
• Find proper training data that will be relevant, like land rights record in different countries we will initially target. (US, Germany, Netherlands, India)
• Research, download and store the relevant training data

Acceptance criteria:

• Identified data sources for land registry information and established mechanisms for data collection.
• Data collection processes automated and running smoothly.

Dependencies:

• Access to Data Sources: Availability of access to land registry data sources is required for data collection.
• Database Infrastructure: Setup and configuration of the database infrastructure are necessary for data storage.

Risks and Mitigation:

• Data Source Availability: Ensure backup data sources are available in case of unavailability or issues with primary data sources.
• Data Integrity: Implement data validation checks and error handling mechanisms to ensure data integrity during collection and storage processes.

Timeline:

• Duration: 12 weeks
• Start Date: 01-12-2024
• End Date: 28-02-2025

Milestone costs

• 25K ADA

Overview:

This milestone focuses on implementing backend infrastructure and services to support the deployment and execution of AI models. It involves setting up the necessary backend architecture, integrating AI models, and developing APIs for model inference and data processing.

Backend Architecture Setup

• Design and set up backend infrastructure to support AI model deployment.
• Select appropriate technologies and frameworks for backend development.
• Ensure scalability, reliability, and security of the backend architecture.

AI Model Integration

• Integrate AI models into the backend infrastructure.
• Develop mechanisms for model loading, initialization, and execution.
• Ensure compatibility and interoperability with different types of AI models.

API Development for Model Inference

• Design and implement APIs for model inference and prediction.
• Define input and output formats for API endpoints.
• Implement error handling and validation mechanisms for API requests.

Acceptance Criteria:

Backend Architecture Setup:

• Backend infrastructure set up and configured according to requirements.
• Technologies and frameworks selected for backend development are suitable for the project needs.

AI Model Integration:

• AI models successfully integrated into the backend infrastructure.
• Mechanisms for model loading, initialization, and execution implemented and tested.

API Development for Model Inference:

• APIs for model inference and prediction developed and documented.
• APIs demonstrate proper input validation and error handling.

Timeline:

Duration: 7 months

Start Date: 01-03-2025

End Date: 30-08-2025

Milestone execution cost

250K

Overview:

This milestone focuses on testing and integrating AI models into the system. It ensures that the models perform accurately and efficiently within the application environment

Acceptance Criteria

Test Environment Setup

• Set up a dedicated test environment for AI model testing.
• Install necessary tools and frameworks for testing.
• Define testing protocols and standards.

Unit Testing

• Develop and execute unit tests for individual components of the AI models.
• Verify the functionality and behavior of each component.
• Ensure that unit tests cover all critical aspects of the models.

Integration Testing

• Integrate AI models with the application or platform.
• Conduct integration tests to ensure seamless interaction between models and other system components.
• Verify data flow and communication channels between the models and the rest of the system.

Performance Testing

• Evaluate the performance of AI models under various conditions (e.g., different input data volumes, concurrent requests).
• Measure response times, throughput, and resource utilization.
• Identify and address performance bottlenecks.

User Acceptance Testing (UAT)

• Invite end users or stakeholders to participate in User Acceptance Testing (UAT).
• Validate the functionality and usability of AI models from the user's perspective.
• Gather feedback and address any issues or concerns raised during UAT.

Timeline:

• Duration: 8 weeks
• Start Date: 01-09-2025
• End Date: 01-12-2025

Milestone execution cost

40K

Overview:

This milestone focuses on documenting the AI model to ensure clarity, transparency, and ease of understanding for stakeholders, developers, and users. Documentation plays a crucial role in explaining the model's functionality, architecture, and usage.

Model Architecture and Functionality Documentation

• Document the architecture of the AI model, including its components, layers, and flow of data.
• Describe the functionality of each component and how they contribute to the overall model's behavior.
• Include diagrams, charts, and illustrations to visualize the model's structure and operations.

Usage and Integration Documentation

• Provide guidelines and instructions for using the AI model, including input data formats, parameters, and output interpretation.
• Document integration steps for developers to integrate the model into applications or systems.
• Include code examples, API documentation, and troubleshooting tips for seamless integration.

Timeline:

• Duration: 3 months
• Start Date: 01-12-2025
• End Date: 02-03-2026

Milestone execution cost

50K

[RESOURCES] Who is in the project team and what are their roles?

• Antonio Hernandez-Garduno: Platform lead in Smart contract development who is a PhD graduate from Caltech, leading security solutions.
• Tony Monicke: Grad student Computer Science, Berlin and experience in the area of generative AI
• Nilay Saha: Worked in leading multinational companies such as ASML, AUDI, Hilti and extensive experience in banking. Has 25+ years of development, solutions architecture and technical leadership experience in large/mid size industry. Also a stake pool operator in Cardano running LKBH pool since the start of Shelley.
• Huy, Vinh, Minh: Developers working on the platform (frontend and backend)
• Giang: UI/UX lead
• Manas Dewan: He is currently the Co-Founder of Brand Kiln Pvt. Ltd. which offers full-suite Metaverse solutions in addition to Consulting, eLearning & Brand Solutions. He was earlier the Chief Operating Officer for the AJP Group, responsible for global brands like Shelby cars, Scomadi scooters & AJ Performance Studio. Previously, Manas worked with the US$ 5 billion OTE Group in Oman where he was Chief Marketing Officer for the Group, responsible for 25+ global brands across Manufacturing, Logistics, Mining, Automotive (Hyundai, Cadillac & 6 auto brands), Consumer products (LG, Samsung, Bosch, & 10+ brands), etc.
• Mauricio Prieto : CSO of Reitcircles Extensive experience beyond blockchain technology and LATAM Cardano DRep .
• Malay Saha: Head of marketing and CEO of Reitcircles. He has been country lead of automotive operations for Isuzu, Hyundai, Carrier Transicold, and has more than 27+ years of marketing experience and leadership.
• Advisors:
• Andrew Westberg: An active member of the Cardano community and has provided key technical advice since the beginning of the project.
• Documentation writer: Has to be hired.

[BUDGET & COSTS] Please provide a cost breakdown of the proposed work and resources

MileStone 1: Data Source Identification and Collection Setup

Amount: 25K

Milestone 2: Backend implementation and architecture

• Amount: 230K

Milestone 3:

• Testing and integration : 20K

Final milestone:

• Usage and overall Documentation: 25K

Total: 300K ADA

[VALUE FOR MONEY] How does the cost of the project represent value for money for the Cardano ecosystem?

We are a bootstrapped project trying to execute on a very lean budget, using hourly market rates for all of the persons concerned. Over time we have developed a set of good processes internally that enables us to roll out a professional product that meets the highest goal of the industry.