rentruewang/aioway

AI on the way. An auto deep learning pipe dream. An RDBMS approach to deep learning. Declarative, explainable, scalable, optimizable, easy to deploy, all that good stuff.

What it solves

Aioway addresses the high entry barrier for machine learning by replacing expensive expert training and subpar AutoML solutions. It aims to overcome the limitations of current tools, which are often too slow (like neural architecture search), inflexible, or non-explainable (like LLMs).

How it works

Inspired by optimizing compilers for SQL, Aioway treats machine learning models and algorithms as instructions. It analyzes the task and available resources to select the best algorithms and models, building a pipeline that remains a "white box" for explainability.

Who it’s for

Developers and data scientists who need an automated, flexible, and explainable way to build ML pipelines without requiring deep expert-level tuning or expensive pretrained models.

Highlights

  • Rule-based optimization: Faster than neural architecture search because it avoids backtracking.
  • Resource-aware selection: Automatically detects tasks and available resources to pick the best model.
  • Explainable architecture: Produces white-box models for better transparency.
  • Flexible interfaces: Supports both a Python library interface and a relational algebra (SQL-like) interface.
  • Scalable: Allows for upgrading model parts and scaling across different machine sizes.

Related

  • Project
    generative-computing/mellea

    A Python library for building predictable AI workflows by turning typed functions into structured LLM calls with automatic validation and retries.

  • Project
    iree-org/iree

    IREE is an MLIR-based compiler and runtime that lowers machine learning models to a unified IR for efficient deployment across data centers, mobile, and edge devices.

  • Project
    zenml-io/zenml

    An open-source MLOps platform that orchestrates the full lifecycle of AI pipelines and agents, from development to production deployment.

  • Project
    WecoAI/aideml

    AIDE ML is an open‑source Python library that implements an LLM‑guided tree‑search agent for automatically writing, debugging, and optimizing machine‑learning pipelines. Users describe a dataset, a goal, and an evaluation metric in natural language; the agent iteratively generates code, evaluates it, and prunes the search tree until the metric is maximized. The package includes a CLI, a Streamlit UI, HTML visualisation, and works with any OpenAI‑compatible LLM (including local Ollama models).

  • Project
    OpenDCAI/DataFlow

    DataFlow is an open‑source Python framework that lets you build, run, and share low‑code pipelines for generating, cleaning, and evaluating LLM training data. It provides a library of reusable operators, a visual WebUI, an AI‑assistant for auto‑creating pipelines, and a Ray‑based distributed execution layer.