Last updated 7 months ago
Developers expect vibe-coding tools when building any app — but popular AI models aren’t trained on Cardano code so vibe-coding for Cardano is impossible, preventing fast iteration and marketability
Integrate our no-code platform with a custom-trained AI model that understands our Cardano building blocks, enabling anyone to create apps that interact with the Cardano ecosystem
This is the total amount allocated to Vibe-coding Cardano Apps in No-code Environment.
Please provide your proposal title
Vibe-coding Cardano Apps in No-code Environment
Enter the amount of funding you are requesting in ADA
78000
Please specify how many months you expect your project to last
10
Please indicate if your proposal has been auto-translated
No
Original Language
en
What is the problem you want to solve?
Developers expect vibe-coding tools when building any app — but popular AI models aren’t trained on Cardano code so vibe-coding for Cardano is impossible, preventing fast iteration and marketability
Does your project have any dependencies on other organizations, technical or otherwise?
Yes
Describe any dependencies or write 'No dependencies'
Yes, the project relies on our existing no-code platform, qod.ee, which is currently in closed beta. We’ve already developed a connector that enables blockchain interactions, along with visual building blocks for integrating and working with a browser wallet. While this approach significantly lowers the barrier to entry compared to traditional coding—and is much faster than building everything from scratch—developers have increasingly gravitated toward vibe-coding tools, preferring them over no-code building blocks. Our next step is to integrate a custom AI agent mode trained to use these building blocks, combining the best of both worlds: a tool that supports vibe-coding while still allowing non-developers to make edits to the AI’s output. The qod.ee platform will offer a free plan so everyone can get started, along with paid tiers for advanced features, more server power, and other enhancements. The AI component for the Cardano model will be credits-based, with a small number of free credits each month. Since AI credits are costly, we won’t be able to support production apps on the free tier, but users will be able to purchase extra credits at prices in line with other AI providers.
Will your project's outputs be fully open source?
No
License and Additional Information
In the first phase, the project will not be open-source, since it depends on the underlying platform. However, we are considering making only the Cardano connector open-source in the future, enabling extensibility and custom plugins from the community.
Please choose the most relevant theme and tag related to the outcomes of your proposal.
AI
Describe what makes your idea innovative compared to what has been previously funded (whether by you or others).
This idea blends a no-code platform custom Cardano connector (powered by https://qod.ee) with a new “vibe-coding” approach, using a custom trained AI agent mode. Unlike other no-code tools that target non-developers or developer-centric platforms that focus on coding, this project seeks to combine the accessibility of drag-and-drop building blocks with the flexibility and speed of "vibe-coding" developer-oriented tools. Adding a custom AI agent trained to use these blocks creates a hybrid experience that everyone can feel comfortable with, regardless of their technical level
Describe what your prototype or MVP will demonstrate, and where it can be accessed.
• Live blockchain integration: Users can connect to multiple Cardano browser wallets, trigger transactions, and read on-chain data through visual blocks
• Usable no-code + AI workflow: A working AI agent can build workflows using the blocks, then allow users to review and tweak the AI’s output
• Real-time editing and preview: Changes can be tested and previewed as they’re made, with clear feedback from both the AI and the visual interface
• The project will be accesible directly on https://qod.ee with users taking advantage of the underlying platform too
Describe realistic measures of success, ideally with on-chain metrics.
Please describe your proposed solution and how it addresses the problem
AI Agent Mode for the Cardano Custom Connector on https://qod.ee
This project will introduce a custom fine-tuned AI model designed exclusively for the Cardano blockchain connector within the existing qod.ee no-code platform. The new AI agent, specifically tailored for Cardano integration tasks, will leverage all existing visual blocks, workflows, and the general Cardano connector.
The unique combination of an AI agent mode with chat-based building allows users to interact naturally while the AI assembles workflows. After the AI makes changes, users are not left with hard-to-read code but with easy-to-use, understandable building blocks that they can further modify as needed.
How It Works:
Unique Value Proposition
Checkout a demo of the current cardano connector here: https://youtu.be/0laVp0eYeTY
Note: For users wanting to try the existing platform, please follow the instructions on the site for joining the closed beta waitlist. Please also mention you're coming from the Ctalyst community.
Also here's a few screenshots:




https://i.imgur.com/Va7fmkA.png
https://i.imgur.com/0rDxH1W.png
https://i.imgur.com/sNivM8e.png
https://i.imgur.com/CxxXrey.png
Please define the positive impact your project will have on the wider Cardano community
This project has the potential to make a significant impact on the Cardano community by dramatically lowering the barriers to building and deploying blockchain applications. By enabling even a few app builders—both developers and non-developers—to quickly create and launch Cardano-integrated apps through the AI-powered connector, the platform can catalyze widespread adoption. Each successfully onboarded builder has the capacity to reach and serve many users, multiplying the adoption and real-world use of Cardano technology. This creates an exponential effect, where the growth of decentralized applications and blockchain interactions expands rapidly through network effects, fueling greater innovation, engagement, and value within the Cardano ecosystem. Ultimately, empowering this new wave of creators will accelerate Cardano’s journey toward becoming a mainstream blockchain platform.
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?
Feasibility
The qod.ee platform already exists, is being actively tested in closed beta right now, and includes functional Cardano building blocks. On top of that, AI technology is widely accessible and proven, with custom training / fine-tuning methods that ensure successful integration of the AI agent.
Capability
I bring over 10 years of experience building complex enterprise applications. The existing qod.ee SaaS platform and the already delivered Cardano connector demonstrate a strong track record of successfully completing complex projects. The incredibly complex no-code platform further highlights this technical expertise.
Milestone Title
Training Data Preparation and Fine-Tuning First Version of AI Model
Milestone Outputs
An independent AI model trained or fine-tuned to understand Cardano connector building blocks and how to combine them. At this stage, the model is standalone—not plugged in the platform connector, can be accessed via an API.
Acceptance Criteria
The model is accessible through an API with a single endpoint for asking questions (prompts)
Model responds accurately to inquiries about known building blocks and how it would combine existing building blocks using the json format; demo effectively showcases model capabilities.
Evidence of Completion
A video or small demo showcasing the AI model’s responses to questions about its capabilities and knowledge of building blocks.
Delivery Month
3
Cost
22000
Progress
30 %
Milestone Title
Integration of AI Model into the cardano connector for Individual Workflows and Pages
Milestone Outputs
The AI model plugged into the qod.ee connector, able to autonomously generate individual workflows or pages based on training after receiving the user prompts. Prompts should be interpreted and executed by the platform, generating individual pages or workflows using the Cardano building blocks. No full agent mode available yet, only individual components.
Acceptance Criteria
Successful generation of valid workflows or pages by the integrated model;
Demo demonstrates clear AI interaction, user prompts should be generating individual pages or workflows using the Cardano building blocks;
Evidence of Completion
Video or small demo of AI model integrated into the connector, showing generation of individual workflows/pages inside the UI platform.
Delivery Month
3
Cost
22000
Progress
60 %
Milestone Title
Implementation of Full Agent Mode with Complete Feature Control
Milestone Outputs
AI model upgraded to full agent mode with comprehensive knowledge of the Cardano connector’s capabilities. The agent can independently create or edit full features, add new pages and/or workflows or make changes to existing ones as needed according to user prompts/requirements. Agent mode is basically similar to vibe-coding tools that work on making a series of changes according to features requested in the user prompt, rather than creating or changing individual components.
Acceptance Criteria
AI agent accurately interprets prompts to generate or edit full features;
Authority to add pages and workflows can be tested and validated;
Demo confirms autonomous feature creation, over multiple pages, workflows, components, not only on a single component. (If the prompt feature request is complex enough to require it)
Evidence of Completion
Video or demo of the AI agent creating and modifying complete features autonomously based on user input
Delivery Month
3
Cost
22000
Progress
90 %
Milestone Title
Public Launch & Documentation
Milestone Outputs
Initial rollout in closed beta for Catalyst users;
Public release of Cardano AI agent mode announced (2 weeks after closed beta started if feedback is good and no big issues);
User documentation; tutorials and support channel ready
Acceptance Criteria
AI Agent available for closed beta users and release date announced;
Ai Agent can be tested and works according to documentation;
Documentation live and accessible;
Support channel live and can be tested with test request;
Evidence of Completion
Launch announcement; user documentation;
AI Agent is accessible in the platform;
Statistics and metrics showing users/number of apps created
Delivery Month
1
Cost
12000
Progress
100 %
Please provide a cost breakdown of the proposed work and resources
Milestone 1: Data Preparation & AI Model Fine-Tuning
Development:
AI Infrastructure (cloud & compute):
Other Costs:
Total: 22,000 ADA
Milestone 2: AI Integration for Individual Workflows/Pages
Development:
AI Infrastructure (cloud & compute):
Other Costs:
Total: 22,000 ADA
Milestone 3: Full Agent Mode Implementation
Development:
AI Infrastructure (cloud & compute):
Other Costs:
Total: 22,000 ADA
Milestone 4: Public Launch & Documentation
Development:
AI Infrastructure (cloud & compute):
Marketing & Community Engagement:
Other Costs:
Total: 12,000 ADA
Any costs exceeding the funding will be supported by me personally or from the profits of the underlying platform.
How does the cost of the project represent value for the Cardano ecosystem?
This project offers great value by focusing on helping the core of the ecosystem: Cardano app developers. They can now iterate faster and reach their markets faster, which increases the overall number of apps in the ecosystem.
So the biggest advantage is the potential for exponential growth. As more app builders use the project’s AI features, their apps attract more users, creating a network effect that drives rapid user growth. This boosts the project’s overall value over time.
With an efficient development approach, costs stay inline with industry standards, ensuring every ADA is well spent. For example:
Hourly rates for developers account for roughly 25ADA average which is below industry standard for seniors in this field (https://www.upwork.com/hire/software-developers/cost/ )
While for testers it averages to 5ADA hourly, which is also way below industry averages. (https://www.upwork.com/hire/software-testers/cost/ )
Terms and Conditions:
Yes
Marian Filipovici - Founder, Architect and main developer of qod.ee (LinkedIn: https://www.linkedin.com/in/marian-filipovici-b1000191/))