Metadata-Version: 2.5
Name: reliax-sdk
Version: 0.0.1
Summary: Per-decision reliability envelopes for AI systems: conformal coverage, calibrated risk brackets, drift tripwires and signed audit records.
Project-URL: Homepage, https://reliax.io
Project-URL: Source, https://github.com/reliax/reliax-python
Project-URL: Issues, https://github.com/reliax/reliax-python/issues
Project-URL: Evaluation, https://github.com/reliax/reliax-evaluation
Author: Reliax
License-Expression: Apache-2.0
License-File: LICENSE
Keywords: ai-governance,calibration,conformal-prediction,drift-detection,model-risk,uncertainty-quantification,venn-abers
Classifier: Development Status :: 1 - Planning
Classifier: Intended Audience :: Developers
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: Intended Audience :: Science/Research
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Topic :: Scientific/Engineering :: Artificial Intelligence
Classifier: Typing :: Typed
Requires-Python: >=3.9
Description-Content-Type: text/markdown

# reliax

**Per-decision reliability envelopes for AI systems.**

> **Status: placeholder release.** This version reserves the package name and
> contains no client API yet. Watch
> [reliax-python](https://github.com/reliax/reliax-python) for the SDK.

Reliax wraps any scoring model and assesses each prediction at inference time,
without touching the model's weights or retraining it. Every decision carries a
formal coverage guarantee, a calibrated risk bracket, a drift verdict, a
routing decision under a policy you own, and a hash-chained audit record an
auditor can replay.

## What is guaranteed

Three components are theorems, and routing runs on them:

| Component | Guarantee |
|---|---|
| Conformal prediction set | contains the true outcome at least 1−α of the time, marginally and per declared segment |
| Venn-Abers bracket | a calibrated probability interval at every score level, from a logit, a probability or scorecard points |
| Test martingale | anytime-valid drift detection; false alarms bounded over the whole run |

## What is not

Two components are signals, labelled as signals, and never enter a routing
rule: the triage rank, and the per-cell calibration opinion. They are monitors,
not routers.

Coverage is a property of the procedure over exchangeable data, not a
probability about any individual prediction. Distribution-free per-instance
conditional coverage is provably unattainable, and we do not claim it.

## Evidence, including the negative results

Measured on public credit data and reproducible end to end at
[reliax-evaluation](https://github.com/reliax/reliax-evaluation). Every
published number is regenerated by a runner script in that repository; nothing
is hand-typed.

The unflattering results are in the same README as the good ones. Under severe
distribution shift our fusion score performs worse than random referral, and it
says so there rather than in a footnote.

## Licence

Apache-2.0. The client SDK, the certificate specification and the verifier are
permissively licensed, permanently — an auditor must be able to verify a Reliax
certificate without anyone's permission.

The reliability engine is licensed separately and is free for non-production
use. See [reliax.io](https://reliax.io).
