From completions to full-lifecycle assistance
GitHub Copilot has grown well beyond the autocomplete suggestions it debuted with in 2021. Today's Copilot Chat in the Visual Studio family handles explanations, refactoring, optimization, debugging, documentation, and test generation. Inline suggestions turn comments into code, fix errors, and write tests for functions as you type.
Interaction with Copilot has expanded beyond plain prompts. Developers can use keywords, slash commands, and chat variables to steer context. Recent additions include support for multimodal models and voice input directly to Copilot.
Perhaps the biggest change is model choice. In Visual Studio Code, Copilot users can pick models from the Azure AI model catalog, including Anthropic's Claude 3.5 Sonnet, Google's Gemini 1.5 Pro (coming soon), and OpenAI's o1-preview and o1-mini. OpenAI and Claude 3.5 Sonnet are available now, at the time of this writing.
The tool's scope now extends into security and review workflows. Copilot Autofix in GitHub Advanced Security detects vulnerabilities in pull requests and generates fixes before merges. AI-powered code review surfaces pull request feedback in editor context. Through Copilot Extensibility, developers can connect Copilot Chat to services like Jira and Sentry, or build custom extensions that pull in their own data and services. A Visual Studio Code extension for GitHub Copilot Workspace also lets developers kick off task-focused work with natural language descriptions of what they want to accomplish.
Deeper integration in Visual Studio
Visual Studio is the designated IDE for .NET, C++, and game development — often against large, multi-project codebases. Copilot integration here includes whole-codebase indexing and search, multi-file edits through the upcoming Edits feature, and debugging and profiler insights. For instance, Copilot works with the Visual Studio profiler to identify and suggest fixes for performance problems, something Microsoft demonstrated on the profiler's own codebase.
Codespaces for instant setup
Developer workstation setup has historically been a chore. The GitHub Codespaces extension for Visual Studio Code bridges local and cloud-based development environments, letting developers create, manage, work in, and delete Codespaces. Both use dev containers as a common format, a fact that helped GitHub cut its own dev box onboarding from 45 minutes to under one minute. Codespaces have also become the standard environment across Microsoft Learn training modules for learning Microsoft technologies, samples, and workshops.
Better collaboration and migration paths
Software is a team sport, and many teams using Azure DevOps have wanted tighter coupling with GitHub. Two products now support that: GitHub Advanced Security for Azure DevOps brings advanced scanning to Azure Repos, and GitHub Copilot for Business brings AI assistance without leaving current workflows.
Integration work also spans Azure Boards, Azure Pipelines, and GitHub Repositories. Developers can create GitHub branches from Azure Boards, link Azure Boards work items in GitHub pull requests, and trigger Azure Pipeline jobs at pull request time or after merges. The goal is end-to-end traceability comparable to Azure DevOps, while allowing teams to tap GitHub's AI and security features. The GitHub Enterprise Importer simplifies moving repos from Azure Repos to GitHub Repositories while keeping Azure Boards and Pipelines active.
Clearing the path to the cloud
Cloud deployment is getting the same integration treatment. The public preview of GitHub Copilot for Azure builds on Copilot Chat in Visual Studio Code to help developers manage resources and deploy applications without leaving the editor. It offers personalized guidance for provisioning, deployment, diagnosis, and cost estimation across Azure services.
Azure CI/CD workflows get tighter GitHub integration
GitHub Actions for Azure now supports workflows for building, testing, packaging, releasing, and deploying to Azure, with built-in utilities for Azure Resource Manager templates, Azure CLI, and Azure Policy. These tools are designed to streamline testing and deployment of intelligent applications using various AI models.
The Azure Developer CLI (azd) provides a direct path in the development loop—provisioning Azure resources, deploying, and setting up CI/CD pipelines for continuous deployments. GitHub Actions workflows are now automatically generated when creating apps with azd templates and infrastructure for popular services like Azure App Service and Static Web Apps.
To help accelerate AI app development, the AI app templates gallery offers a curated collection of deployable templates for Visual Studio Code or GitHub Codespaces. The templates prioritize reliability, repeatability, and security, and support a wide range of programming languages, frameworks, models, and services—including options from vendors like Pinecone, Arize, LangChain, and LlamaIndex.
Tools for building intelligent applications
Keeping pace with the rapidly evolving AI field can be challenging. GitHub Models enables secure experimentation with Azure AI models, offering QuickStart code across various languages and frameworks to simplify model exploration. Developers can move from experiment to prototype to production by transitioning from GitHub Models to Codespaces, Visual Studio Code, and Azure. Those unsure where to start can experiment in the GitHub playground for free, then scale AI apps to paid endpoints using Azure for enterprise-level deployment and monitoring.
Azure AI evaluation and online experimentation Actions (both in preview) allow scaling of AI applications with key insights. These capabilities integrate into CI/CD workflows by posting pre-production evaluation and online experimentation results directly back to GitHub for analysis.
Self-provisioning GitHub from the Azure portal
GitHub can now be purchased and self-provisioned directly from the Azure portal, including setting up an Enterprise Managed Users configuration. GitHub Enterprise Cloud with data residency is built on Azure infrastructure, providing the security, business continuity, and disaster recovery capabilities inherent to the cloud.
Microsoft and GitHub remain committed to supporting developers worldwide in their innovation and collaboration efforts, combining strengths to shape the next generation of technology.
