CHAPTER 1: PROJECT INTRODUCTION & MARKET CONTEXT
1.1. Project name and vision:
Project name: “CoinWise AI”
Meaning: A combination of "Coin" (Digital Asset), "Wise" (Intelligence), and "AI"
(Artificial Intelligence).
Our mission: To become the leading digital financial education platform for Gen Z, helping
them enter the digital asset market with confidence, knowledge, and absolutely no capital
risk.
1.2. Market context analysis:
The explosion of Digital Banking:
The banking industry is shifting strongly towards the "Open Banking" model and integrating
new asset classes. Crypto is no longer a foreign concept but is gradually becoming a part of
the investment portfolios of digital banking users.
The wave of young investors:
In Vietnam and Southeast Asia, the percentage of young people (Gen Z & Millennials)
owning digital assets is among the highest in the world. However, the majority of them
participate due to herd mentality (FOMO) without a basic knowledge base.
1.3. Identifying the problem (Pain points):
The project focuses on addressing the three biggest barriers for new users:
- Financial Risk: The digital asset market is extremely volatile. A small mistake at the
beginning can lead to real financial losses, causing negative psychological effects.
- Technical Barrier: The interfaces of professional trading platforms (Binance, OKX)
are too complex with hundreds of indicators, easily discouraging newcomers.
- Education Gap: Financial literacy is often dry and uninteresting. Young people need a
learning environment that combines theory with practice, but it must also be safe.
1.4. Target audience:
Demographics: Ages 18-30 (Students and recent graduates).
Behavioral characteristics:
- They own smartphones and are familiar with mobile banking/electronic wallet
applications.
- They are interested in online investment options.
- They prefer minimalism and modernity, and need immediate support from technology
(AI).
Motion development: "CoinWise AI is not designed to replace exchanges, but rather to act as
an 'educational sandbox' within the digital banking ecosystem, helping users practice before
transitioning to real-world financial services."
CHAPTER 2: BUSINESS MODEL
2.1. Value propositions:
CoinWise AI delivers core value based on three pillars:
- Risk-Free system: Users are provided with virtual funds for practice, completely
eliminating the risk of losing real money during the learning process.
- Real-time experience: Utilizing real data from Binance, this simulation provides a
100% realistic experience that closely mirrors actual market fluctuations.
- AI-Driven Mentorship: Gemini AI assistants act as personal consultants, explaining
complex terminology and analyzing trends instantly for each individual.
2.2. Revenue streams strategy:
The project adopts the Freemium model (basic free and premium fees apply) – a popular
model in current digital banking applications:
Free package Trade 4 major coin pairs, with a Free (Using to attract users).
virtual capital of $1.000.000, and
AI chat limited to 5 messages per
day.
Premium package Unlock the entire Altcoin Subscription fee applies
portfolio, unlimited AI chat, and (~$4.99/month)
weekly market analysis reports.
2.3. Competitive advantage:
Compared to older trading apps or physical exchanges, CoinWise AI has unique advantages:
- Minimalism (Lean UX): Eliminate complex features like Margin and Futures trading,
focusing only on Spot trading so that beginners are not overwhelmed.
- Closed ecosystem: Integrated deposit flow (via emulated Stripe) and transaction
history to fully simulate a real banking application.
- Integrated AI technology: Displays a chat window with AI in case users need answers
to questions related to finance in general and trading in particular.
2.4. Cost structure analysis:
To operate this model, the main costs include:
- Technology costs: API maintenance fees (Gemini/Stripe) when the user base is large.
- Infrastructure costs: Hosting and user data storage.
- Marketing costs: Attracting Gen Z users through social media and simulated trading
contests.
CHAPTER 3: USER EXPERIENCE (UX/UI DESIGN)
3.1. Design language:
CoinWise AI opts for a modern design style, focusing on the Gen Z audience:
- Dark Mode: This is the industry standard for financial and crypto applications (it
helps users focus on charts, reduces eye strain, and creates a "high-tech" feel).
- Minimalism: Use cards and white space appropriately to separate the following areas:
Charts - Wallet - Trading Orders.
3.2. User experience flow:
The system is designed so that a novice can make their first trade in less than 2 minutes:
- Onboarding: Register quickly → Receive $1.000.000 in virtual capital immediately.
- Education: The AI assistant suggests questions to help users understand more directly
on the home screen.
- Deposit (Simulation): Experience the actual deposit process via Stripe to understand
how funds flow into the system.
- Trading: Select coin → View real-time chart → Click "Buy" → Monitor portfolio.
- Arena Feature:
+ The system creates an Arena, where multiple accounts can participate
simultaneously in a time-limited trading simulation.
+ All participants trade using real-time market data, ensuring fair competition.
At the end of the session, users are ranked, and the top performer receives the
reward.
+ This feature enhances engagement and competitiveness, transforming
simulated trading into an interactive, game-based learning experience suitable
for FinTech education and events.
3.3. Critical UX points:
A. Candlestick charts:
Instead of using static charts, the project integrates the Lightweight Charts library. This gives
a professional feel, like using a real investment banking tool, allowing users to interact, zoom
in/out, and view past prices.
B. Visual feedback:
- Loading states: When calling APIs from Binance or Stripe, the buttons will display a
spinning effect (spinner). This is extremely important in digital banking so that users
know the system is processing and avoid repeatedly pressing buttons, which could
cause errors.
- Success/error modes: Use dynamic dialog boxes (modal) instead of the crude alert()
command. Each success message is encouraging, making learning more engaging for
the user.
C. Portfolio dashboard: Displays the total net asset value, continuously updated according
to market prices. This helps users develop psychological resilience by seeing their assets
fluctuate every second.
CHAPTER 4: ENGINEERING ARCHITECTURE & SYSTEM LOGIC
4.1. The information flow diagram:
To ensure real-time performance, CoinWise AI operates in a closed loop process from
external data sources to the user interface:
- Data Source: Directly connected to the Binance API.
- Intermediate layer (CORS Proxy): Uses a proxy to bypass browser security barriers,
allowing JSON data to be retrieved from international servers to the local
environment.
- Data processing: Extracting price parameters (open, high, low, close) to pass into the
charting library.
- Rendering: Charts and price tables automatically update after each cycle (1-3
seconds) without needing to reload the page.
4.2. The trading engine logic:
This is the core part that simulates the operation of a digital bank/exchange. The logic is built
on the principle of "Test first - Execute later":
*Buy Order:
Condition: USD balance >= Current price x Quantity.
Processing: Deduct cash → Add assets to wallet → Record in journal.
*Sell Order:
Condition: The number of assets in the wallet must be greater than or equal to the number of
assets you wish to sell.
Processing: Deduct assets → Add cash (calculated at market price at the time of sale) →
Record in journal.
→ Portfolio Update: The system continuously recalculates the Gross Net Asset Value (NAV)
using the following formula:
NAV = Cash balance + Total (Quantity of assets[i] x Instantaneous market price[i]).
4.3. Data persistence:
Even without using a traditional backend to optimize resources for the MVP, the project still
ensures data is not lost thanks to the following techniques:
- LocalStorage: Acts as a local database storing account information, balances, and
portfolios.
- JSON Serialization: Converts complex data structures into text strings for storage and
vice versa, helping to maintain login state and assets even when the user refreshes the
page (F5) or closes the browser.
CHAPTER 5: INTEGRATING BREAKTHROUGH TECHNOLOGIES (AI &
PAYMENT)
5.1. Stripe payment integration:
In a digital banking system, the cash-in process is the most important. CoinWise AI integrates
the Stripe SDK to simulate this process:
- Security: Instead of processing card information itself, the system uses Stripe's
Iframe. This ensures compliance with PCI DSS standards (card data is not stored
directly on the project server).
- User experience: Using test mode to simulate the response from the card-issuing
bank. Insert card information → Wait for verification → Receive transaction result
notification.
- The logic states: Cryptocurrency is only credited to the account after Stripe returns the
success_token code, ensuring the integrity of financial data.
5.2. AI Virtual Assistant (Gemini AI Integration):
The biggest differentiating factor of the project is transforming chatbots into true data
analytics experts:
- Prompt engineering: The system not only sends user questions but also includes
market context.
- Real-time analysis: Thanks to its connection with the Large Language Model (LLM),
the AI assistant can explain complex candlestick chart fluctuations that beginners
cannot understand on their own, shortening the learning process.
5.3. Information security & authentication: Despite being a simulated platform, CoinWise
AI still applies the Session Management security principles of Digital Banking. This includes
using sessionStorage to automatically log users out when they close their browsers,
preventing unauthorized access on public computers.
CHAPTER 6: RISK MANAGEMENT & COMPLIANCE
6.1. Regulatory compliance:
In the context of a gradually evolving legal framework for digital assets, CoinWise AI is
designed to adhere to the basic sandbox principle (experimental mechanism): the project acts
as an "educational buffer zone". Because it uses virtual capital and has no real-world
exchange value, the project does not fall under the category requiring complex financial
exchange business licenses, thus minimizing initial legal risks.
6.2. Simulating the anti-fraud process:
Although it's a simulation, to standardize the Digital Banking process, the project
incorporates a Transaction Monitoring feature. This system stores the entire transaction
history (Ledger). In reality, this data will be scanned by AI algorithms to detect unusual
behavior (such as price manipulation or money laundering), in accordance with international
standards.
6.3. Operational risk management:
The project establishes contingency plans for technical incidents:
- Risk of data loss: Use a periodic backup mechanism to localStorage. In the future, the
system will synchronize to the cloud (Cloud Sync) to ensure that user data is not lost
when changing devices.
- Risk of API data inconsistencies: When the Binance API or Gemini API experiences
issues, the system will automatically switch to "Offline Mode" or use the most recent
cached data to avoid disrupting the user experience.
- Client-side security: Simple data encryption in localStorage prevents users from
arbitrarily modifying virtual balances through the browser's Inspect Element tool.
CHAPTER 7: CONCLUSION & FEASIBILITY ASSESSMENT
7.1. MVP assessment:
The project has successfully built a logically complete prototype for a next-generation digital
banking application:
- In terms of technology: Successfully integrating real data (Binance), real payments
(Stripe), and artificial intelligence (Gemini) onto a single platform.
- In terms of user experience: A smooth user flow has been established, from
registration and deposit to transactions and history tracking.
- Regarding education: Solving the problem of "risk-free practice" and creating
practical value for the young investor community.
7.2. Current limitations:
To maintain the objectivity of the product, we need to acknowledge the limitations of the
current product.
- Data storage: Using localStorage is only suitable for individual use; the system needs
a centralized database (Backend) to manage thousands of concurrent users.
- Security: The API key is currently located on the frontend and needs to be moved to
the server-side environment to ensure absolute security.
- Features: Currently focused only on Spot trading, not offering other complex financial
products such as Savings (Staking) or Lending.
7.3. Development roadmap:
CoinWise AI is more than just a simulator. Some of its next development roadmap includes:
- Phase 1 (Scale-up): Migration to a Fullstack architecture ([Link]/MongoDB),
integration with the Firebase authentication system.
- Phase 2 (Social Trading): Develop a simulated "Copy Trade" feature, allowing users
to monitor and learn strategies from players with the highest profit margins. -
- Phase 3 (Institutional Partnership): Connect with banks to become an official
"Financial Learning" subsystem within their existing Mobile Banking applications.
7.4. Conclusion:
The CoinWise AI project is a testament to the fact that the combination of Digital Banking
and AI can remove barriers to financial literacy. By prioritizing user experience and security,
we believe this model has high commercialization potential and will make a positive
contribution to improving financial literacy in the digital age.