Wolfram Plutus RAG: The AI-Powered Knowledge Engine

[Proposal setup] Proposal title

Please provide your proposal title

Wolfram Plutus RAG: The AI-Powered Knowledge Engine

[Proposal Summary] Budget Information

Enter the amount of funding you are requesting in ADA

40000

[Proposal Summary] Time

Please specify how many months you expect your project to last

5

[Proposal Summary] Translation Information

Please indicate if your proposal has been auto-translated

No

Original Language

en

[Proposal Summary] Problem Statement

What is the problem you want to solve?

Plutus knowledge is scattered, hard to search, and often unreliable. Developers need a system that generates precise, citation-backed answers from curated Plutus resources.

[Proposal Summary] Supporting Documentation

Supporting links

• https://www.wolfram.com/notebook-assistant-llm-kit/
,
• https://writings.stephenwolfram.com/2024/07/yet-more-new-ideas-and-new-functions-launching-version-14-1-of-wolfram-language-mathematica/#rags-and-dynamic-prompting-for-llms
,
• https://wolframblockchainlabs.com

[Proposal Summary] Project Dependencies

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

No

Describe any dependencies or write 'No dependencies'

No dependencies

[Proposal Summary] Project Open Source

Will your project's outputs be fully open source?

Yes

License and Additional Information

We intend to license the project outputs under the Apache License 2.0, aligning with Cardano Foundation standards. Apache 2.0 chosen for patent protection and commercial compatibility while ensuring community can freely use, modify, and distribute. We believe this strikes a good balance between encouraging collaboration and protecting the project's integrity.

[Theme Selection] Theme

Please choose the most relevant theme and tag related to the outcomes of your proposal

AI

[Campaign Category] Category Questions

Mention your open source license and describe your open source license rationale.

We intend on utilizing Apache License 2.0, chosen specifically for its alignment with Cardano ecosystem standards and enterprise adoption requirements. This license provides explicit patent grants protecting contributors and users while allowing commercial use cases that ensure project sustainability.

How do you make sure your source code is accessible to the public from project start, and people are informed?

Code will be on a public GitHub from project start, with initial commits covering project details and documentation. We will post weekly updates in GitHub Discussions, Discord, and the Catalyst forum.

How will you provide high quality documentation?

This project will include a clear README, developer setup guides with screenshots and code snippets. The RAG pipeline will include examples and video walkthroughs to demonstrate the workflow from start to finish.

[Your Project and Solution] Solution

Please describe your proposed solution and how it addresses the problem

Plutus is the foundation of Cardano’s smart contract ecosystem, enabling secure, functional-programming-based decentralized applications. However, despite its capabilities, Plutus suffers from a documentation access problem:

• Official docs are scattered across multiple sites and repositories
• Cardano Improvement Proposals (CIPs) are stored in GitHub with limited discoverability
• Key best practices are buried in community discussions and forum threads.
• Resources quickly become outdated as Cardano evolves.

Developers, auditors, and researchers are forced to manually gather information from scattered and inconsistent sources, slowing development and deterring new contributors.

The Plutus RAG project will build a Retrieval-Augmented Generation (RAG) AI system designed to be the authoritative technical knowledge engine for Plutus. It will unify and curate all relevant Plutus content into a single, searchable system and pair it with an AI assistant that provides accurate, context-aware, citation-backed answers.

To guarantee reliability, Plutus RAG will draw exclusively from vetted, authoritative materials, including:

• Official Plutus and Cardano documentation
• Cardano Improvement Proposals (CIPs)
• GitHub repositories, commits, and code samples
• Cardano Foundation resources
• Developer forums, mailing lists and archived discussions.
• Peer-reviewed academic papers related to Plutus and UTXO models

We will aggregate and preprocess all Plutus-related documents, enriching them with metadata such as source, topic, and date to enable precise and efficient searching. This curated content will then be stored in a vector database, allowing for advanced semantic retrieval so queries can be matched by meaning, not just keywords.

When a user asks a question, the system will retrieve the most relevant context and pass it to a large language model, which will generate natural-language answers complete with in-line citations and links to original sources.

The backend will also expose REST APIs for seamless integration into IDEs, governance platforms, and other external workflows. Acting as an open-book reference for Plutus, this system will significantly improve accuracy, and strengthen smart contract security. Built with future expansion in mind, the RAG system will also support onboarding of additional data sources to keep pace with the evolving Cardano ecosystem.

[Your Project and Solution] Impact

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

Plutus RAG addresses a major challenge in the Cardano ecosystem by resolving the problem of scattered, inconsistent, and outdated information about Plutus. By consolidating official documentation, CIPs, community guides, and technical discussions into a single authoritative reference, it will make accurate, verified information instantly accessible, enabling faster, more secure smart contract development and auditing.

Expected Outcomes

• Developer Efficiency: Cut search time for precise answers from hours to seconds, letting developers focus on building and optimizing contracts.
• Security Improvements: Reduce bugs and vulnerabilities by providing up-to-date, contextually accurate information aligned with the latest Plutus releases and best practices.
• Audit Readiness: Give auditors instant access to compliance references, transaction logic explanations, and source-backed validations to accelerate assessments.
• Ecosystem Growth: Lower onboarding barriers for new developers, reducing learning time and improving retention.
• Research Enablement: Offer structured, searchable resources for academic and industry research into Plutus programming and security.

[Your Project and Solution] Capabilities & 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?

Wolfram’s competitive edge lies in the scale, precision, and maturity of its data curation and computational infrastructure, exemplified by the Wolfram|Alpha Computational Knowledge Engine and the Wolfram Knowledgebase, one of the largest curated computational datasets in the world. This infrastructure integrates thousands of authoritative sources, including government records, scientific databases, financial markets, and live data feeds.

The ingestion pipeline processes vast volumes of structured and unstructured content, applying rigorous normalization, consistency checks, and metadata enrichment to ensure data is always current and computation-ready. This capability, refined over decades, enables high-precision analytics, reasoning, and dynamic query generation across domains, predating and extending beyond the generative AI era.

Building on this expertise, Wolfram can create tailored data pipelines and knowledge architectures for Cardano’s Plutus. By integrating automated ingestion and enrichment with an LLM-powered retrieval-augmented generation (RAG) conversational layer, the system can enable semantic search and context-aware knowledge retrieval.

Organizational Strengths

• Our team has over three decades of leadership in computational science, data science, and AI. We're used to supporting innovative systems around the world.
• We have multidisciplinary team specializing in AI, blockchain development, web development, data science, and community engagement.
• Our company has always prided ourselves on the global talent and knowledge we have. Along with our innovative tech stack, we have an in-house consulting team who has helped organizations with the most difficult problems for decades.

[Milestones] Project Milestones

Milestone Title

Plutus RAG Architecture, GitHub launch

Milestone Outputs

Core Plutus RAG system architecture defined and documented (data ingestion, vector DB, LLM integration, API layer). GitHub repository initialized with base code structure and Apache 2.0 license. Initial README with architecture diagrams and setup instructions.

Acceptance Criteria

Architecture will document and cover all core components and interfaces. The GitHub repository will be public with initial commits. A local environment setup will be verified using instructions in the README documentation.

Evidence of Completion

A gitHub repository link with initial commits and README file will be included. An architecture diagram will be included in the repo docs folder.

Delivery Month

1

Cost

5000

Progress

10 %

Milestone Title

Data Aggregation & Preprocessing Pipeline

Milestone Outputs

Aggregation of all identified Plutus-related sources (official docs, CIPs, GitHub repos, Cardano Foundation resources, forums, etc.). Preprocessing scripts for cleaning, normalizing, and adding metadata to documents. Vectorized dataset stored in preparation for vector database ingestion.

Acceptance Criteria

All listed sources are retrieved and stored in a raw AND processed formats in order to make both convenient use and also fact checking.

Scripts will be run successfully and generate searchable datasets for the verifier.

Evidence of Completion

GitHub repository updated with preprocessing scripts and metadata schema.

Dataset sample files with documented fields.

Delivery Month

2

Cost

10000

Progress

40 %

Milestone Title

Semantic Retrieval Engine & Vector Database Setup

Milestone Outputs

A vector database will be configured for semantic search on Plutus documents.

An embedding pipeline will be integrated for document indexing and retrieval.

And search API endpoints for semantic queries will be implemented.

Acceptance Criteria

A vector database will be operational with the ingested processed dataset.

Queries will return relevant ranked results for efficient verification.

And a search API will be documented with example requests and responses.

Evidence of Completion

GitHub repository updated with vector database configuration and search API source code. Demo video showing semantic search on indexed Plutus content.

Delivery Month

3

Cost

10000

Progress

60 %

Milestone Title

LLM Integration & Answer Generation

Milestone Outputs

A RAG pipeline integrated with a selected LLM to generate natural language answers. An in-line citation system will link back to original the sources in order to provide a clear understanding of what informed the answers.

Acceptance Criteria

The AI assistant via RAG returns context-aware answers which will support the desired output. A demonstration of the working system should show that citations consistently map to the correct document and section.

Evidence of Completion

GitHub updated with the with RAG + LLM integration modules. Sample Q&A logs with source references will be provided.

Delivery Month

4

Cost

10000

Progress

80 %

Milestone Title

REST API Integration, Documentation, Community Testing, Final Release

Milestone Outputs

REST API for external integration with IDEs, governance platforms, and workflows. Developer and user documentation. Video tutorials and examples. Community testing feedback incorporated into the final release.

Acceptance Criteria

A REST API passes acceptance testing forr automated and manual functional tests with adequate documentation able to be verified by Cardano community. REST API for external integration with IDEs, governance platforms, and workflows.

Evidence of Completion

GitHub repository containing final API code, documentation, and developer tools. Postman collection for API endpoints.

Delivery Month

5

Cost

5000

Progress

100 %

[Final Pitch] Budget & Costs

Please provide a cost breakdown of the proposed work and resources

The total project budget is ₳40,000, allocated to ensure each development stage is fully resourced, tied to measurable outputs, and aligned with the 5-month delivery plan.

Plutus RAG Architecture & GitHub Launch – ₳5,000

Funds the definition of the core system architecture, environment setup, and the creation of a public GitHub repository licensed under Apache 2.0. Includes architecture diagrams, initial README documentation.

Data Aggregation & Preprocessing Pipeline – ₳10,000

Covers the aggregation of authoritative Plutus sources (official docs, CIPs, GitHub repos, Cardano Foundation resources, and community discussions), along with preprocessing scripts for cleaning, normalizing, and adding metadata. Outputs a vector-ready dataset for ingestion into the semantic search engine.

Semantic Retrieval Engine & Vector Database Setup – ₳10,000

Funds the configuration of the vector database and the integration of the embedding pipeline for semantic search. Includes API endpoints for ranked, meaning-based document queries, with example requests/responses and a working search demo.

LLM Integration & Answer Generation – ₳10,000

Supports the integration of a retrieval-augmented generation pipeline with a large language model to produce context-aware, citation-backed answers. Includes an in-line citation framework and testing to ensure accuracy and source traceability.

REST API Integration, Documentation & Final Release – ₳5,000

Enables the delivery of the final REST API for integration with IDEs, governance platforms, and external workflows. Funds developer and user documentation, video tutorials, community testing, and incorporation of feedback into the final release.

[Final Pitch] Value for Money

How does the cost of the project represent value for the Cardano ecosystem?

The Plutus RAG initiative delivers exceptional value by combining Wolfram’s proven expertise in large-scale knowledge curation, retrieval-augmented generation (RAG), and computational reasoning with targeted, high-impact outcomes for the Cardano ecosystem.

The requested budget for Plutus RAG delivers both immediate and lasting value to the Cardano ecosystem:

• By reducing Plutus information search time from hours to seconds, developers can focus on building, testing, and deploying smart contracts, resulting in measurable productivity gains.
• The RAG framework, ingestion pipeline, and vector database can be extended beyond Plutus to other Cardano languages.

By aligning cost with direct, high-impact benefits, Plutus RAG not only addresses an immediate developer need but also establishes a scalable foundation for ongoing innovation and ecosystem expansion.

[Required Acknowledgements] Consent & Confirmation

Terms and Conditions:

Yes