Metadata-Version: 2.4
Name: aidflow
Version: 3.0.0
Summary: Pipeline execution framework
Author-email: AIGENSYS <info@aigensys.com>
Requires-Python: >=3.13
Description-Content-Type: text/markdown

# AIDflow 

Most ML tools are simple wrappers around Scikit-Learn—they automate the process but hide the logic, creating a "black box" that regulators reject. AIDflow is different.

AIDflow is designed specifically for high-stakes environments (Fintech, Banking, Insurance) where a model isn't just a file—it's a legal document. 
We provide a full ecosystem for scorecard development that prioritizes *Human-in-the-Loop Governance*:

*Absolute Micromanagement*: While the pipeline is automated, every single step is transparent and open. Users can pause, intervene, and fine-tune any parameter at any layer—from raw ingestion to final champion selection.
*Audit-Ready Reporting*: Every phase generates comprehensive, industry-standard reports. We don't just give you an accuracy score; we provide the evidence required for internal risk committees and external regulators.
*Production Scorecard Readiness*: The output is not just a .pkl file—it is a fully validated production scorecard, complete with stability metrics and explainability vectors that meet the strictest global financial standards.
*The Governance Gap*: AIDflow fills the void between "fast prototyping" and "compliant deployment," ensuring your models are as robust in the eyes of a regulator as they are in a Jupyter notebook.

## Why this is a winning strategy:
It solves the "Fear of Automation": By using words like "Micromanagement" and "Human-in-the-Loop," you tell the bank's risk officer that they are still in control.
It targets high-value niches: Mentioning "Regulators," "Scorecards," and "Financial Industry" immediately moves your product from a "hobbyist tool" to a "mission-critical enterprise asset."
It defines the category: You aren't competing with Scikit-Learn; you are creating a new category: Regulatory Scorecard Orchestration.