The applied-ml repository is a curated collection of papers, articles, and blog posts documenting how companies implement data science and machine learning systems in production environments. Rather than theoretical research, the repository focuses on practical case studies from major technology organizations, showing the real-world decisions, techniques, and outcomes behind deployed ML systems.
The repository is organized into 31 distinct topic areas covering the full spectrum of production ML work. These sections range from foundational concerns like data quality and data engineering, through core ML techniques including classification, regression, forecasting, and recommendation systems, to specialized domains such as computer vision, natural language processing, reinforcement learning, and graph-based approaches. Additional sections address infrastructure concerns including model management, MLOps platforms, validation and A/B testing, and efficiency optimization. The repository also includes sections on team structure, ethical considerations, privacy-preserving techniques, and documented failures, providing a comprehensive view of production ML beyond just successful implementations.
Each topic section contains curated links to specific company blog posts, research papers, and technical talks, typically with publication dates and company attributions. For example, the data quality section includes contributions from Airbnb, Uber, Google, Amazon, Gojek, and Netflix, each describing their approaches to ensuring data reliability at scale. The data engineering section similarly documents frameworks and platforms from companies like Airbnb, Netflix, DoorDash, and Uber, showing how organizations structure their data pipelines and feature engineering infrastructure.
According to GitGenius tracking data, the repository has experienced modest but steady growth, with fork count increasing from 3964 to 3965 between July 2026 checks. The repository maintains minimal issue and pull request activity, with a median response latency of approximately 1963.7 hours across tracked items. The most active contributors tracked by GitGenius include Dommati-Sahithi and Huaian666, each with single recorded events. The repository shares contributors with major open-source projects including Microsoft's VSCode and TypeScript implementations, as well as the Rust language project, suggesting it attracts contributors from significant technology communities.
The repository serves as an educational resource for practitioners implementing ML systems, offering insights into how established companies frame problems, select techniques, and measure success. By aggregating real-world case studies rather than creating original content, the repository provides a structured index into the distributed knowledge of production ML practice across the technology industry. The breadth of topics and the emphasis on both successes and failures make it a reference point for understanding contemporary approaches to deploying machine learning at scale.
