DAID

• 

  Framer Setup

• 

  Landing Page

• 

  Framer Output

• 

  Solution Design Output

• 

  Solution Design Setup

Inspiriation

Small business owners and startup founders often suffer from "analysis paralysis.". When faced with challenges such as fluctuating market demands or scaling operations, they usually rely on "gut feel" or informal advice. Traditional consulting is prohibitively expensive, and static business templates are too rigid to adapt to the specific nuances of their business. They need a way to apply professional-grade strategic rigor without the overhead of a consultant.

What it does

DAID is a decision intelligence tool designed for small business owners and entrepreneurs. It transforms messy business thoughts into structured, actionable execution plans using consultant like business framing and battle tested decision frameworks. It has two sections: Framer and Solution Design. The Framer helps the user to structure their thought into a structured format. This can be refined as many times as needed. Solution Design leverages a plethora of decision making frameworks to help the user make a smarter decision. It can work two ways: the first way is tt take the updated information from the framer. The second is that the user can enter their problem here and then use solution design to come up with the best answer to their problem. It also has the ability to self select the right framework if the auto-select option is chosen. This is all powered by the airia platform.

How it was built

It was built using FastAPI, airia platform and Vercel.
FastAPI was selected due to its native asynchronous architecture allows it to handle long-running AI analysis requests without freezing the user interface.
Vercel was used to host the web app.
Airia platform was used for AI orchestration. It took user input from the web app and generated smart output using an AI model

Design

User Experience

Users interact with DAID via a streamlined interface. They can start with framer and then go to the Solution design or go straight into the solution design.

AI Layer

DAID moves beyond simple analysis. It has a framer agent and a solution design agent.
The Framer agent acts as a Strategic Consultant. It evaluates the user's input, determines if sufficient information exists to perform a high-quality analysis. it deconstructs the problem into its fundamental truths.
The solution design agent also evaluates the user's input, determines if sufficient information exists to perform a high-quality analysis, and it can use a selected framework or choose the best framework based on its analysis when "auto-select" is chosen by the user. It doesn't just build an execution plan from the ground up, it provides a level of structural rigor previously unavailable to small-scale operations. It doesn't just analyze a problem; it architecturally solves it. In terms of design both agents use the Input --> AI Model --> Output design and were built and tested in airia studio.
For the AI model, I chose Claude Haiku 4.5. I wanted to output to be succint but focused. Kept the temperature at around 0.2 and reasoning effort balanced for both agents. Designed a unique output schema in json based on the output I needed for both agents. API information and keys were easily generated and stored secretly.

Deployment

The app was deployed on Vercel. It was a smooth process as it works seamlessly with FastAPI. I also able to embed all the airia API credentials easily.

Challenges

The top challenge was researching different decision making frameworks. I did not expect there to be so many out there. I did have some issues with the airia url for the different agents but experimented with different approaches until I found a workable solution.

Learning Experience

It was a fun experience learning about the different decision frameworks. I also learned more about the airia platform and what it offers from a design and busienss perspective.

Next Steps

In terms of what is next for the app, I plan on adding these functionalities

• I will add more decision making frameworks.
• Add the ability to choose different languages
• Add authentication and multi-user ability
• Add ability to save outputs
• Ability to analyze combine text and image inputs

Built With

• airia
• fastapi
• vercel