AI is being used to augment the software delivery system of GitHub through the deployment of multi-model orchestration, autonomous coding agents, and automated code review.
AI is now being deployed to augment the software development system of GitHub by means of autonomous coding, multi-model orchestration, and automated verification as developers rely more on delegating software tasks to autonomous agents. This represents a paradigm shift that has been happening within the development of software for electronics systems.
These new features comprise GitHub Copilot Fusion, Copilot Projects, Copilot Automations, and Copilot Code Review Max. Fusion involves selecting and coordinating a number of AI models depending on the task at hand. These AI models may involve reasoning, code generation, debugging and using tools. The goal of this strategy is to strike a balance between computational power, execution time and cost.
Copilot Projects is a feature that supports multiple sessions of agents working in parallel in multiple repositories, while session continuity lets developers handle cloud tasks using Visual Studio Code, command line, GitHub app and website. Copilot Automations are capable of running scheduled and event-based workflows such as code reviews and issue management, and learn from past actions and provide suggestions for improvement. Code Review Max identifies severe defects in software prior to merging.
Evidence of this is shown in the Octoverse Report of 2026 by GitHub. There were 17.9 billion code pushes to GitHub over the last year, with a total of 34.3 billion commits. In September 2026, more than 40% of pull requests were done by AI agents, compared to 17% in March. Also, more than 61% of Copilot users selecting models selected several models during their coding.
However, the transition is not only taking place in software companies but extending even to manufacturing, healthcare, agriculture and industries, where developers use AI-assisted tools to solve engineering problems specific to their domains. To support this trend, GitHub has provided four million CPU cores, 160 PB of high-speed storage and 50 Tb/s of network bandwidth. In addition to this, GitHub is moving to multi-regional Azure environment and rearchitecting the Git system to improve its scalability, reliability and throughput for autonomous development workloads.
