T3 Research Intelligence

What the company appears to be: T3 Research Intelligence is a self-reported research and decision-support system for retail investors in Vietnam’s stock market. It combines Wyckoff principles, quantitative analysis, T3 market signals, cited web research, and private holdings into a structured workflow. The author states it was built during OpenAI Build Week as an extension to a pre-existing personal system.

What changed: The project description indicates that the author, James Do, developed a new "Shared Web Research and Portfolio Intelligence layer" during Build Week, distinct from prior work. This new layer connects public market research with private portfolio data through a strict boundary enforcement mechanism.

Single most important open question: Is there any evidence of real-world usage or adoption by retail investors beyond the author’s personal system?

Analysis basis: The entire analysis is based on the self-reported, unverified project description provided by the caller. No external verification, revenue, customer data, or traction metrics are available.

The description states that T3 Research Intelligence is a portfolio intelligence system for retail investors in Vietnam’s stock market. It integrates:

• Wyckoff principles
• Quantitative analysis
• T3 market signals
• Cited web research
• Private holdings

It connects three layers:

1. T3 market context – market, sector, and security signals.
2. Shared research – structured public-market research tied to a specific T3 data batch and reusable across workflows.
3. Private portfolio analysis – user-scoped holdings, cost basis, profit and loss, plans, and interpretation.

The system enforces a strict boundary between shared and private data: “Shared research must never contain private portfolio data.” This is enforced via mutation tests, deep scans, and fail-closed controls.

Claim: The product is described as a research and decision-support system.

Evidence: Self-reported in the project write-up.

The author states that he built this tool because he lacked a repeatable decision-making process when entering the Vietnamese stock market. He wanted to answer not just “what happened in the market,” but “what changed, why does it matter to the positions I hold, and can I trace the evidence behind the analysis?”

This suggests an evolution from emotional trading to disciplined, evidence-based investing.

The system is positioned as a tool for retail investors, particularly those who struggle with manual research and portfolio management.

Claim: The product aims to support disciplined decision-making.

Evidence: Self-reported in the “Why I built it” section.

The description states that T3 Research Intelligence is intended for retail investors in Vietnam’s stock market. It was initially built as a personal system and later considered for broader use by other retail investors.

It targets users who:

• Lack structured decision-making processes
• Struggle with manual research across multiple sources
• Want to integrate public market data with their private holdings

Claim: The target customer is a retail investor in Vietnam.

Evidence: Self-reported, no external validation.

There is no evidence of pricing, monetization strategy, or business model in the description. The author states that the system was built for personal use and may be made available to others if it proves useful.

Claim: No pricing or business model described.

Evidence: Not evidenced.

The system is built using:

• AI tools (GPT-5.6, Codex, Claude Code)
• Frontend: React, Next.js, TailwindCSS
• Backend: Node.js, Python, SQLite
• Data pipeline: T3 market signals, external research providers
• Architecture: Fail-open for availability, fail-closed for integrity and privacy

Key technical features include:

• Shared research reuse across workflows
• Private data isolation via mutation tests and schema validation
• Resolver sequence: memory cache → durable snapshot → join in-flight request → call provider → validate and normalize → persist → publish
• Automated test suite with 696 tests (695 passed, 0 failed)

Claim: The system uses AI-assisted development and enforces strict data boundaries.

Evidence: Self-reported.

There is no evidence of traction or adoption beyond the author’s personal use. The system has not been verified in production with real providers, and the author explicitly states that “the public-production feature remains disabled until controlled validation succeeds.”

The project was submitted to a hackathon (OpenAI Build Week), but there is no indication of any commercial deployment or user base.

Claim: No traction or maturity signals.

Evidence: Not evidenced.

No competitive landscape or market positioning is described. The author does not mention competitors or similar tools in the Vietnamese stock market space.

Claim: No competitive context provided.

Evidence: Not evidenced.

• Unverified production use: The system has not been validated with real providers, and the public feature remains disabled.
• No commercial traction: No evidence of users beyond the author.
• Self-reported only: All claims are unverified.
• AI-assisted development without independent audit: While Codex was used for testing, no third-party verification is mentioned.

Inference: The system may not yet be ready for production use or widespread adoption.

Evidence: Self-reported, no external validation.

1. What are the actual performance metrics of real-provider integration (latency, hit rate, persistence success)?
2. How is the private data boundary enforced in practice? Has it been tested under real-world conditions?
3. Is there any evidence of user feedback or adoption beyond personal use?
4. What is the plan for scaling beyond Vietnam’s stock market?
5. Are there any plans to monetize or commercialize the system?

Inference: These questions are necessary due to lack of external validation and production data.

Evidence: Not evidenced.