Uploaded August 2026 | Updated November 2026, 3 days ago
Optimize GitHub Copilot context and token usage to get better AI results with fewer tokens. Learn how the context window works and use /context, /init, and /compact to keep Copilot sessions focused, accurate, and efficient.
Episode 5 of 10 in Build your personal brand with Copilot. Continue to Episode 6 and explore the full series:
https://aka.ms/student-learning-series-website
Build your personal brand with Copilot is a beginner-friendly Microsoft Developer series for students. Follow Microsoft employees as they use GitHub Copilot, Visual Studio Code, GitHub Pages, Microsoft Foundry, MCP servers, and agent skills to create, publish, and improve a professional portfolio website.
In this episode:
0:00 Welcome and an introduction to context and tokens
0:21 What context includes
0:38 How Copilot processes information as tokens
0:46 Viewing context with /context
0:53 Understanding the context window with a suitcase analogy
1:04 Why too little or too much context can hurt results
2:07 Generating a case study page without project context
3:04 Adding project files for better results
3:57 Refreshing project context with /init
4:02 How copilot-instructions.md guides Copilot
5:02 Starting a focused session with the right context
6:08 Checking context usage
6:44 Reducing token usage with /compact
7:11 Recap of context and token optimization
Connect with the presenters:
Emily Rothrock - Software Engineer at Microsoft: linkedin.com/in/emily-rothrock
Shivani Sundaresan - Software Engineer at Microsoft: linkedin.com/in/shivanisundaresan
Resources:
Download the prompts, code, and episode artifacts from GitHub: https://aka.ms/student-learning-series
Learn more about GitHub Copilot: https://aka.ms/StudentAI-GitHubCopilotApp
Some features and cloud services may have eligibility, usage, or cost limits. Review the current terms before creating resources.
Subscribe to Microsoft Developer for more practical developer and AI learning:
youtube.com/@MicrosoftDeveloper
#GitHubCopilot #PromptEngineering #AITools
Optimize GitHub Copilot context and token usage to get better AI results with fewer tokens. Learn how the context window works and use /context, /init, and /compact to keep Copilot sessions focused, accurate, and efficient.
Episode 5 of 10 in Build your personal brand with Copilot. Continue to Episode 6 and explore the full series:
https://aka.ms/student-learning-series-website
Build your personal brand with Copilot is a beginner-friendly Microsoft Developer series for students. Follow Microsoft employees as they use GitHub Copilot, Visual Studio Code, GitHub Pages, Microsoft Foundry, MCP servers, and agent skills to create, publish, and improve a professional portfolio website.
In this episode:
0:00 Welcome and an introduction to context and tokens
0:21 What context includes
0:38 How Copilot processes information as tokens
0:46 Viewing context with /context
0:53 Understanding the context window with a suitcase analogy
1:04 Why too little or too much context can hurt results
2:07 Generating a case study page without project context
3:04 Adding project files for better results
3:57 Refreshing project context with /init
4:02 How copilot-instructions.md guides Copilot
5:02 Starting a focused session with the right context
6:08 Checking context usage
6:44 Reducing token usage with /compact
7:11 Recap of context and token optimization
Connect with the presenters:
Emily Rothrock - Software Engineer at Microsoft: linkedin.com/in/emily-rothrock
Shivani Sundaresan - Software Engineer at Microsoft: linkedin.com/in/shivanisundaresan
Resources:
Download the prompts, code, and episode artifacts from GitHub: https://aka.ms/student-learning-series
Learn more about GitHub Copilot: https://aka.ms/StudentAI-GitHubCopilotApp
Some features and cloud services may have eligibility, usage, or cost limits. Review the current terms before creating resources.
Subscribe to Microsoft Developer for more practical developer and AI learning:
youtube.com/@MicrosoftDeveloper
#GitHubCopilot #PromptEngineering #AITools


![VS Gets AI Agents + MCP C# SDK 2.0 and More
Welcome to your weekly roundup of some Microsoft news in 2 minutes.
The gist:
Microsoft shipped an agent preview inside Visual Studio built on the same
GitHub Copilot SDK that powers Copilot CLI, plus Foundry model updates and
MCP C# SDK 2.0. Heres the Microsoft Developer news worth paying attention
to this week in The Upload.
Links for more:
https://aka.ms/devblogs.microsoft.com/ai
https://aka.ms/devblogs_visual_studio_update
https://aka.ms/devblogs_foundry_update
In this episode:
→ Visual Studio agent preview — built on the GitHub Copilot SDK behind
Copilot CLI, with built-in Azure and .NET skills authored by Microsoft
engineering teams. Designed for larger tasks with less back-and-forth.
→ Why Copilot is converging on one agent platform across Visual Studio,
VS Code, GitHub, and the CLI — a consistent agent experience on every surface.
→ Microsoft Foundry updates — [MODEL NAME: TBD] image generation,
[MODEL NAME: TBD] transcription, more model options, and expanded
tooling for agent development.
→ Why the advantage is shifting from access to a model to access to a
complete AI platform.
→ MCP C# SDK 2.0 — the latest MCP spec, stateless-first communication,
and improved interoperability.
→ Why MCP is becoming the integration layer for AI agents the way APIs
were for everything before.
📺 Subscribe for a new Upload every week. VS Gets AI Agents + MCP C# SDK 2.0 and More](https://i.ytimg.com/vi/4aX8cI15-Ng/mqdefault.jpg)







