https://github.com/avatariamfoundation-afk/neurogrid-core

https://github.com/avatariamfoundation-afk/neurogrid-core

blockchain consent desci medical opbnb rpm

Last synced: 5 months ago
JSON representation

• Host: GitHub
• URL: https://github.com/avatariamfoundation-afk/neurogrid-core
• Owner: avatariamfoundation-afk
• License: mit
• Created: 2025-12-12T20:14:01.000Z (6 months ago)
• Default Branch: main
• Last Pushed: 2026-01-06T22:29:43.000Z (5 months ago)
• Last Synced: 2026-01-07T07:30:53.138Z (5 months ago)
• Topics: blockchain, consent, desci, medical, opbnb, rpm
• Language: Solidity
• Homepage:
• Size: 408 KB
• Stars: 0
• Watchers: 0
• Forks: 0
• Open Issues: 0
• Metadata Files:
  • Readme: README.md
  • License: LICENSE

Awesome Lists containing this project

README

          
# NeuroGrid  

**Deterministic Medical Compute Coordination Layer**

---

## Executive Summary

**NeuroGrid** is a deterministic, auditable coordination layer for medical-grade computation, artifact validation, and telemetry governance.  

It is designed to support **trust-minimized medical AI workflows** without performing diagnosis, treatment, or autonomous clinical decision-making.

NeuroGrid provides:

- Deterministic compute registration and execution

- Verifiable artifact emission and lineage

- Fault-coded telemetry and slashing enforcement

- DAO-governed policy control

- Hackathon-safe token abstraction with a clear post-hackathon transition path

This repository represents the **Core Protocol Layer** only.

---

## What NeuroGrid Is (and Is Not)

### NeuroGrid **IS**

- A **coordination protocol** for medical compute

- A **governance and accountability framework**

- A **verifiable execution substrate**

- A **compliance-aware infrastructure layer**

### NeuroGrid **IS NOT**

- A diagnostic system

- A clinical decision engine

- A medical device

- An autonomous AI practitioner

> NeuroGrid does not replace clinicians, regulators, or medical judgment.

---

## Core Objectives

1. **Determinism**  

   Every compute, artifact, and decision is reproducible and traceable.

2. **Accountability**  

   Validators, compute nodes, and governance actors are fault-bound and slashable.

3. **Safety-First Design**  

   No opaque inference. No hidden execution. No silent failure.

4. **Hackathon Compliance**  

   Uses placeholder token naming while preserving real-world viability.

---

## Architecture Overview

NeuroGrid is composed of **modular, auditable smart contracts** and off-chain compute nodes governed by deterministic rules.

### On-Chain Components

- **NeuroGridKernel** – Global state coordination

- **ComputeRegistry** – Compute node registration and lifecycle

- **ArtifactRegistry** – Artifact issuance and lineage tracking

- **ValidatorManager** – Validator enrollment, scoring, and slashing

- **NeuroDAO** – Governance and policy control

- **ProposalExecutor** – Deterministic proposal execution

- **Telemetry Interface** – Fault-coded execution reporting

### Off-Chain Components

- Deterministic compute nodes

- Signed artifact producers

- Telemetry emitters

---

## Deterministic Trust Model

NeuroGrid enforces trust through:

- Immutable fault codes

- Slashing-bound validator behavior

- Explicit failure states

- No implicit trust assumptions

All trust boundaries are documented and enforced on-chain.

---

## Telemetry & Fault Codes

Every execution path emits:

- Deterministic telemetry

- Explicit fault codes

- Verifiable execution outcomes

No silent success.  

No silent failure.

See:

- `FAULT_CODE.md`

- `DETERMINISTIC_SLASHING_POLICY_SPEC.md`

- `TELEMETRY.md`

---

## Token Model (Hackathon-Safe)

### During Hackathon

- **MedToken** (placeholder)

- Non-speculative

- Non-listed

- Used only for governance simulation and validator incentives

### Post-Hackathon (Planned)

- **ABSToken**

- Real-world utility token

- Funded by awarded winnings

- Intended uses include:

  - Validator incentives

  - Governance participation

  - Telemetry verification

  - Ecosystem sustainability

> Token transition is explicitly deferred until after regulatory review and audit.

---

## Security Posture

- No upgradeable proxies without DAO approval

- No privileged execution paths

- Explicit admin role separation

- Defense-in-depth design

See:

- `SECURITY_MODEL.md`

- `DEFAULT_ADMIN_ROLE.md`

- `GRID_ADMIN_ROLE.md`

- `STRUCTURAL_HARDENING.md`

---

## Testing & Validation

The system includes a full test matrix covering:

- Happy paths

- Permission failures

- State-guard violations

- Fault escalation paths

See:

- `TEST_MATRIX.md`

---

## Deployment Notes

NeuroGrid is designed for:

- EVM-compatible chains

- Cross-chain telemetry compatibility

- Deterministic deployment order

See:

- `DEPLOYMENT_NOTES.md`

---

## Regulatory & Ethical Positioning

- Infrastructure-only

- No medical claims

- No patient data handling

- Designed to **support** compliance, not replace it

---

## Judge-Facing Clarity

This submission demonstrates:

- Strong system design discipline

- Deterministic trust enforcement

- Clear separation between infrastructure and clinical use

- Responsible handling of post-hackathon funding intent

Awarded funds will be directed toward:

- Independent medical AI audit

- Legal and IP structuring

- Infrastructure hardening

- Validator network maturation

- Future compliant ecosystem expansion

---

## Repository Status

- Core protocol: **Complete**

- Documentation: **Hardened**

- Tests: **Defined**

- Demo & Interface: **Next Phase**

---

## Final Note

NeuroGrid is intentionally conservative.

It is built to **earn trust before demanding adoption**.

---

**Status:** Core Protocol – Finalized