Awesome Software Engineering for Machine Learning

URL: https://github.com/SE-ML/awesome-seml
Category: Themed Directories
Tags: machine-learning, software-engineering, awesome-lists
License: CC0-1.0

Overview

Awesome Software Engineering for Machine Learning is a curated, awesome-style directory of articles and resources on software engineering practices for production-level machine learning systems. It focuses on the engineering aspects around ML models rather than on developing new ML algorithms, covering topics such as data ingestion, coding, testing, versioning, deployment, quality control, and team collaboration.

The repository also underpins a survey on how software engineering practices are adopted in applications with ML components.

Features

Scope and Focus

• Concentrates on software engineering practices for ML applications, not on core ML algorithm research.
• Emphasizes production-level and operational ML systems.
• Covers the full lifecycle around ML components: data, development, testing, deployment, and maintenance.

Structured Thematic Sections

Resources are organized into clearly defined sections:

• Broad Overviews
  • High-level resources that cover multiple aspects of software engineering for ML.
• Data Management
  • Practices and guidance on data ingestion, preparation, management, and related engineering aspects.
• Model Training
  • Resources on engineering robust, reproducible, and maintainable training pipelines.
• Deployment and Operation
  • Articles and guidelines on deploying ML models, running them in production, monitoring, and operations.
• Social Aspects
  • Materials on collaboration, team processes, and organizational aspects around ML engineering.
• Governance
  • Resources on policies, risk management, and oversight for ML-enabled systems.
• Tooling
  • Information on tools that support software engineering practices in ML projects.

Resource Types and Indicators

• Includes a mix of:
  • Articles and technical blog posts.
  • Reports and guidelines from organizations.
  • Academic and scientific publications.
• Uses simple markers:
  • ⭐ Must-read – highlights particularly important or foundational resources.
  • 🎓 Scientific publication – denotes academic or peer-reviewed work.

Community and Contribution

• Hosted as a public GitHub repository (SE-ML/awesome-seml).
• Includes:
  • code-of-conduct.md – community standards for participation.
  • contributing.md – guidelines for proposing additions or changes to the list.
• Open to community contributions via pull requests and issues.

Related Survey and Reading

• Based on the curated literature, the maintainers offer:
  • A survey on the adoption of software engineering practices in ML-enabled applications: https://se-ml.github.io/survey
  • Additional explanatory material on practices: https://se-ml.github.io/practices

Pricing

• Not applicable. This is an open, CC0-licensed GitHub repository of resources and is free to access and use.