Expectro Health — Clinical Evidence

The company appears to be a single-person project (Claudio Carvalho) building a clinical decision support platform in Brazil using OpenAI technologies and specialized health data sources. The author states the goal is to help physicians access, analyze, and understand relevant clinical evidence through a persistent, traceable conversation with a GPT-powered virtual assistant.

The description states that Expectro Health — Clinical Evidence is a bilingual clinical decision support platform powered by OpenAI functionalities and a specialized evidence retrieval layer. It allows physicians to:

• Initiate a persistent clinical session
• Receive a unique session code
• Retrieve the same session later
• Describe signs, symptoms, medications, test results, and clinical progression
• Retrieve relevant scientific, pharmacovigilance, and regulatory evidence
• Ask follow-up questions to a Virtual General Practitioner using GPT technology
• Preserve the conversation and linked evidence for traceability

The platform is described as built with OpenAI technologies, including Codex and the OpenAI Responses API. It integrates with 13 specialized evidence indexes such as PubMed, NCBI Bookshelf, Anvisa drug data, VigiMed adverse reactions, drugs and notifications, Notivisa, technovigilance, hemovigilance, drug restrictions, pricing, and toxicological data.

The description states that the platform extends the vision of ChatGPT Health with Brazil-specific clinical evidence, pharmacovigilance, and regulatory data—bilingual, traceable, and built for clinicians. The author emphasizes:

• It is not meant to replace physicians but to help them access, analyze, organize, and understand relevant evidence
• The physician remains the ultimate decision-maker
• The platform supports persistent and traceable conversations

This positioning suggests a clinical support tool rather than an autonomous diagnostic or treatment recommendation system. The claim evolution appears to be focused on combining conversational AI with specialized Brazilian health data infrastructure.

The description states that the target customer is clinicians, specifically physicians in Brazil who need access to specialized clinical evidence beyond general health information. The platform is described as built for clinicians and aims to help them access, analyze, organize, and understand relevant evidence while maintaining the physician as the ultimate decision-maker.

Not evidenced. The description does not contain any information about pricing, monetization strategy, or business model.

The platform is built using:

• OpenAI technologies (Codex, GPT, OpenAI Responses API)
• Engineering agent: Codex for auditing repositories, understanding architecture, implementing persistent sessions, creating backups and rollback manifests, building bilingual interface, connecting evidence adapter, implementing OpenAI Responses API layer, creating automated tests, identifying regressions, verifying protected components
• Workflow process: Clinical case → Retrieval of specialized evidence → Evidence linked to the session → GPT explanation → Follow-up conversation → Physician's decision
• Development approach: Small, controlled phases where Codex implements and tests, user decides, project memory records

The platform is described as using RAG (Retrieval-Augmented Generation) with 13 specialized evidence indexes.

Not evidenced. The description states that this was submitted to the OpenAI 2026 hackathon and contains no information about revenue, customers, or adoption beyond the author's own account.

Not evidenced. The description does not contain any information about competitors or market positioning relative to existing clinical decision support tools.

• Single-person team: Only one member listed (Claudio Carvalho)
• Hackathon submission: No evidence of commercial traction or product-market fit beyond a hackathon project
• No revenue or customer data: The description does not contain any information about actual users, customers, or monetization
• Unverified claims: All features and capabilities are self-reported without independent verification
• Limited evidence of clinical adoption: No indication that physicians are actually using the platform beyond the author's own account

1. What is the current status of the platform? Is it in production or still a prototype?
2. Have you had any actual users or clinicians test the platform?
3. How do you plan to monetize this platform?
4. What are your plans for scaling beyond the hackathon submission?
5. Are there any partnerships with Brazilian health institutions or regulatory bodies?
6. What is the timeline for bringing this to market?

evidenced

The description states: "Physicians in Brazil need more than just general health information."

Inferred

The platform is built for clinicians, but the specific roles or specialties are not detailed.

evidenced

The description states: "Our goal is not to replace physicians. The goal is to help them access, analyze, organize, and understand relevant evidence, while maintaining the physician as the ultimate decision-maker."

It also states: "A physician can initiate a persistent clinical session... Retrieve relevant scientific, pharmacovigilance, and regulatory evidence... Ask follow-up questions to a Virtual General Practitioner using GPT technology... Preserve the conversation and linked evidence for traceability."

Inferred

The value proposition is tied to the integration of AI with specialized Brazilian health data, but the specific benefits beyond access are not detailed.

evidenced

The description states: "A physician can initiate a persistent clinical session... Receive a unique session code... Retrieve the same session later."

It also mentions: "We use Codex as an engineering agent to implement persistent sessions; create backups and rollback manifests; build the bilingual interface; connect the evidence adapter; implement the OpenAI Responses API layer..."

Inferred

The method of how physicians access or discover the platform is not specified.

evidenced

The description states: "A physician can initiate a persistent clinical session... Retrieve the same session later."

It also mentions: "The project memory records" and "preserve the conversation and linked evidence for traceability."

Inferred

There is no explicit statement about how customer support or engagement is managed.

not evidenced

The description does not mention any revenue model, pricing, or monetization strategy.

evidenced

The description states: "OpenAI is fundamental to both the engineering process and the clinical experience."

It also mentions: "We use Codex as an engineering agent... GPT feeds the conversational layer through the OpenAI Responses API."

Inferred

The specific resources such as data repositories, APIs, or infrastructure are not detailed beyond their use.

evidenced

The description states: "A physician can initiate a persistent clinical session... Describe signs, symptoms, medications, test results, and clinical progression... Retrieve relevant scientific, pharmacovigilance, and regulatory evidence."

It also mentions: "Codex implements and tests. The user decides. The project memory records."

Inferred

The activities related to development or maintenance of the platform are not fully detailed.

evidenced

The description states: "We use Codex as an engineering agent... GPT feeds the conversational layer through the OpenAI Responses API."

It also mentions: "We work in small, controlled phases: Codex implements and tests. The user decides. The project memory records."

Inferred

There are no explicit partnerships with external organizations or institutions mentioned.

not evidenced

The description does not mention any cost structure, including operational costs, development expenses, or resource allocation.

1. Customer Segments: Marked as evidenced based on the statement "Physicians in Brazil need more than just general health information." The specific roles or specialties are inferred and would require further detail to be evidenced.

1. Value Propositions: Marked as evidenced based on the stated goal of helping physicians access, analyze, organize, and understand relevant evidence. The benefits beyond access are inferred.

1. Channels: Marked as evidenced based on the description of persistent sessions and session retrieval. How physicians discover or access the platform is inferred.

1. Customer Relationships: Marked as evidenced based on the ability to initiate, retrieve, and preserve sessions. Engagement or support mechanisms are inferred.

1. Revenue Streams: Marked as not evidenced; no mention of pricing, monetization, or revenue model.

1. Key Resources: Marked as evidenced based on use of OpenAI, Codex, and GPT. Specific resources like data repositories are inferred.

1. Key Activities: Marked as evidenced based on the described workflow of session initiation and evidence retrieval. Development or maintenance activities are inferred.

1. Key Partnerships: Marked as evidenced based on use of Codex and GPT. Explicit partnerships with external entities are inferred.

1. Cost Structure: Marked as not evidenced; no information provided about operational or development costs.

To convert the inferred blocks into evidenced ones, the following questions would be needed:

• What specific roles or specialties of physicians does the platform target?
• How do physicians discover and access the platform?
• What are the specific benefits beyond access that the platform provides?
• Is there a pricing model or monetization strategy in place?
• What are the specific data repositories or infrastructure used?
• What are the development or maintenance activities involved?
• Are there any explicit partnerships with external organizations or institutions?
• What are the operational or development costs of the platform?