MIRA

The description states that MIRA is:

• A trading terminal with memory
• Not a signal app, robo-advisor, or portfolio tracker
• Built for US, Hong Kong, China, and crypto markets
• Multilingual (English, Simplified Chinese, Japanese)
• Designed to help users make disciplined decisions by integrating:
  • Portfolio data
  • Personal rules
  • Trade history
  • Emotional state
  • AI insights grounded in evidence

It includes features such as:

• A dashboard for net worth, performance, allocation, drawdown, and “trading temperature”
• Holdings, watchlists, and a trade log that records reasons behind actions
• A four-step Position Builder to formalize trade ideas before execution
• Personal rules and allocation targets that check decisions against them
• AI Insights and Copilot that point out risks, opportunities, rule breaks, and repeated behavior
• Trade review and monthly reports for learning from past performance

The product is described as never placing orders or telling users what to buy — the decision remains theirs.

Confidence Low. This is a self-reported feature list with no independent verification of functionality or completeness.

The author claims MIRA:

• Is not just another portfolio tracker
• Is not a signal app or robo-advisor
• Focuses on helping users see themselves, not the market
• Treats personal rules, emotional state, and past behavior as data for AI
• Uses AI to check plans against evidence rather than predict markets

It positions itself as:

• A discipline loop: track → decide → reflect
• An investor’s mirror, showing internal patterns instead of external signals
• A tool that helps users stay responsible and in control

The evolution from a personal journaling skill to a full demo suggests an intent to scale beyond the author’s own use case, but no evidence is provided about how this transition was validated or tested.

Inference The positioning implies a shift from a niche personal tool toward a broader investor-facing product. However, this is inferred from the narrative and not substantiated by any external data.

The description states:

• MIRA targets individual investors
• Works across US, Hong Kong, China, and crypto markets
• Supports three languages: English, Simplified Chinese, Japanese

It is built for users who:

• Invest their own money
• Want to slow down before acting
• Are interested in learning from emotional patterns and repeated behaviors
• Value discipline over market timing or automation

There is no mention of institutional investors, brokers, or other segments.

Confidence Low. No segmentation data or customer personas are provided beyond the author’s personal experience.

The description does not state:

• Any pricing model
• Revenue streams
• Monetization strategy
• Subscription plans or fees

It says:

• MIRA is an MVP built for demonstration
• It uses preset sample data in demo mode
• No real brokerage connections are mentioned yet
• The author intends to add broker integrations and import options

Inference If MIRA moves beyond demo, it may adopt a freemium or subscription model, but this is speculative.

The description states:

• Built with React, Vite, Tailwind CSS, shadcn/ui, and JavaScript/Python backend
• Supports English, Simplified Chinese, Japanese
• Uses Tencent Finance, yfinance, CoinGecko for market data
• Employs Supabase for data handling
• AI layer powered by GPT-5.6 via OpenAI Responses API
• Uses Codex for UI design and product iteration during Build Week

It also mentions:

• Deterministic code checks portfolio, rules, allocation, and behavior before invoking AI
• AI only provides explanations backed by real evidence, never executes trades
• Demo data stays in browser and can be reset

Confidence Medium. The technical stack is detailed, but no information on scalability, security, or production readiness.

The description states:

• MIRA is an early MVP
• Built during a short Build Week hackathon
• Not yet tested for all paths or edge cases
• Still has bugs and data issues outside main flows
• No real brokerage account integration in demo
• No mention of users, customers, or revenue

The description does not provide:

• Names of competitors
• Market share data
• Competitive positioning
• Comparison to existing tools in the space

It implies MIRA is different from:

• Signal apps
• Robo-advisors
• Portfolio trackers

But it does not define its competitive landscape or how it stands out.

1. No traction or user data: The product is described as an MVP built for a hackathon with no real-world testing.
2. Unproven AI utility: While AI is central, there's no evidence that it adds value beyond what deterministic checks could provide.
3. Single-founder build: Only one team member (Lydia Lee) is listed; no indication of scaling capability or team structure.
4. Limited market reach: Currently supports only 3 languages and 4 markets — not a global solution yet.
5. No monetization strategy: No pricing, revenue model, or business plan described.
6. Demo-only functionality: The demo uses sample data; real-world integration is still in development.

1. What specific user problems are you solving that current tools don’t?
2. How do you plan to validate the effectiveness of AI insights with actual users?
3. Are there any early adopters or beta testers using MIRA in real accounts?
4. What is your path to monetization and long-term sustainability?
5. How will you scale beyond the current 3 languages and 4 markets?
6. What are the key assumptions about user behavior that underpin the product design?
7. Have you considered how AI bias or overconfidence might affect decision-making?
8. What are the technical challenges in moving from demo to production-grade data handling?