iints-sdk-python35 1.5.18

A pre-clinical Edge-AI SDK for diabetes management validation.

IINTS-AF SDK

"Code shouldn't be a secret when it's managing a life."

Open-source research platform for insulin delivery algorithm simulation and validation.

---

Repository Map

Path	Purpose
src/iints/	installable SDK source code
docs/	public documentation, governance, release notes, and Obsidian vault mirror
examples/	runnable examples and notebooks
tests/	unit, CLI, system, and realism tests
research/	research artifacts and benchmark snapshots
tools/	maintainer scripts for docs, CI, and release workflows
data_packs/	small bundled/demo data packs
algorithms/	example insulin algorithm entry points

Important starting points:

• docs/THEORY_STRESS_LAB.md — Jetson scientific stress-test mode
• docs/governance/API_STABILITY.md — semver and API policy
• docs/governance/PRIVACY_POLICY.md and docs/governance/TERMS_OF_USE.md — public legal docs

---

What It Does

• Simulate virtual patients across thousands of scenarios before any algorithm reaches a real device
• Certify datasets cryptographically — traceable, consented, reproducible
• Understand results with audit-ready reports and local AI explanation via Ministral

---

Install

pip install "iints-sdk-python35[full,mdmp]"
iints doctor --smoke-run

Raspberry Pi / Arduino UNO Q:

pip install "iints-sdk-python35[edge,mdmp]"
iints edge quickstart --board raspberry_pi
iints edge quickstart --board uno_q

---

Quick Start

iints quickstart --project-name my_study
cd my_study
iints presets run --name baseline_t1d --algo algorithms/example_algorithm.py
iints ai report results/<run_id>

Full documentation: python35.github.io/IINTS-SDK

---

Live Demo

For a Zoom call, jury walkthrough, or sponsor demo where you want the story first and code as proof:

iints demo eucys --output-dir results/live_demo

Use iints demo doctor for clinical feedback conversations and iints demo booth for public digital-patient demos.

---

Research software. Not a medical device. Apache-2.0 licensed, with legacy MIT notices where applicable. Built by a 17-year-old with type 1 diabetes.