README

This repo is put together to help me document, practice and show examples of how to use Copilot. Along the way, if I find authoritative content on the topic of Generative AI, I will list them here. All content is credited to its respective authors.

Now, remember folks! "Copilot still needs a Pilot" (C) 2023

A bit of History

AI > ML > DL > GI

• 1956 Artificial Intelligence (AI) - field of computer science that seeks to create intelligent machines that can replicate or exceed human intelligence
• 1997 Machine Learning (ML) - subset of AI that enables machines to learn from existing data and improve upon that data to make decisions or predictions
• 2017 Deep Learning (DL) - machine learning technique in which layers of neural networks are used to process data and make decisions
• 2021 Generative AI (GI) - a new class of AI that can generate new content, such as text, images, and code

Machine Learning

Predictive models based on data and statistics

1. Anomaly Detection: Systems that detect unusual patterns or events, enabling pre-emptive action
2. Computer Vision: Applications that interpret visual input from cameras, images or videos
3. Matural Language Processing: Applications that can interpret and draw insights from written or spoken language
4. Conversational AI: AI agents or bots that can engage in dialogs with human users

Origins of Machine Learning

• 1950 - 1980's Symbolic AI
• 1990's Machine Learning
• 2010's Deep Learning

What types of problems does machine learning help us solve?

• Spam and fraud detection
• Decision engines: approval or disapproval
• Recommendation engines: which one to choose
• Predictive maintenance: when will it break
• Image analysis: classification and object detection
• Text analysis: term and document comprehension
• Natural language processing and machine translation

Responsible AI Prinicples

• Fairness
• Reliable
• Safe
• Private
• Secure
• Inclusivness
• Transparent
• Accountable

OpenAI + Microsoft Offerings

• Generative pre-trained transformer (GPT): GPT (generate and understand text)
• Codex: (based off of GPT and are optimized to generate and understand code)
• Dall-E: (generate images from text prompts)
• Embeddings: (a special format of data representation that can be easily utilized by machine learning models and algorithms)

Introduction to Copilot by Copilot

Copilot is a new AI-powered coding assistant that helps you write code faster and with fewer bugs.

You can receive suggestions from GitHub Copilot either by starting to write the code you want to use, or by writing a natural language comment describing what you want the code to do.

Trained on billions of lines of public code, GitHub Copilot puts the knowledge customers need at their fingertips, saving time and maintaining focus.

Although it supports most programming languages, it currently works the best with

• Python
• JavaScript
• TypeScript
• Ruby
• Go

Value Prop of Github Copilot

GitHub Copilot helps developers code faster, focus on solving bigger problems.

• 96% faster with repetitive tasks
• 88% feel more productive
• 74% focus on more satisfying work

Copilot Use Cases

• Find new solutions - Cycle through suggestions and discover a different path.
• Solve big problems - Spend less time on boilerplate and repetitive code patterns and more time on building great software.
• Explore new frameworks - Navigate unfamiliar languages, frameworks, and libraries with ease

User Stories: I as a Developer want to... so that...

1. I want to write code faster so that I can focus on solving bigger problems
2. I want to write code with fewer bugs so that I can focus on solving bigger problems
3. I want to write code with fewer errors so that I can focus on solving bigger problems
4. ... LOL. The above was written by my Copilot. I am not sure if I should be happy or sad.

As a Developer

1. You are asked to create a new App
2. You are asked to create a new API
3. You are asked to modify an existing application
4. You are asked to fix a bug

Your next steps might be to gather requriements, work on a high level design, produce a POC or MVP. And when you start to write code, you may copy/paste, you may search and query for best results and code samples, etc. Copilot can help you do it all in one place.

The Formula

Let's say the scenario is: "I want to build a web app that connects to a sql database and has an todo item api and then I want to deploy this app to Azure". Imagine now, with [Copilot, Copilot Labs, Copilot-X, Copilot CLI] or which ever combination or product term we may end up with I can do all of that from one IDE.

1. Context. Context. Context.
2. Well written natural language comments (in a code file) or prompts.
3. Multiple tabs open in the IDE with code files related to the work you are doing.

Copilot Supported IDEs

Copilot supports the following IDEs:

• Visual Studio
• Visual Studio Code
• JetBrains
• NeoVim
• GitHub Codespaces
• GitHub Copilot Workspace

How does it work?

GitHub Copilot is powered by OpenAI Codex from https://openai.com/

Who can get GitHub Copilot

1. GitHub Copilot can be managed through personal accounts with GitHub Copilot for Individuals
2. Or through organization accounts with GitHub Copilot for Business
3. You can get it for Free if you are a verified Student, Teacher, or Maintainer of a popular open source project

High level steps to get started

1. Sign up with a GitHub account
2. Purchase a Copilot license
3. Active on a GitHub repo
4. Install the extensions (Copilot and Copilot Labs)
5. Sign up for Copilot-X
6. Create your first code file and enter a comment. Use natural language

Behind the Scenes

• Prompt Engine - The Prompt Engine is a library for building natural language prompts for code generation. It is used by Copilot to generate prompts for code completion.

• Prompt Engineering - The Prompt Engineering website is a collection of resources for building natural language prompts for code generation. It is used by Copilot to generate prompts for code completion.

• Semantic Kernel - Semantic Kernel is an open-source SDK that lets you easily combine AI services like OpenAI, Azure OpenAI, and Hugging Face with conventional programming languages like C# and Python. In other words, it is an AI orchestration layer that allows us to combine AI models and plugins together to create brand new experiences for users.

• Language Model - A language model is a probabilistic model of a natural language.

Example Applications built ontop of the tech

• AutoGPT
• Copilot Chat

Public Announcements

• Introducing GPT-4o: OpenAI new flagship multimodal model now in preview on Azure

• GitHub Copilot Workspace

• Microsoft Copilot for Azure. Article date. 11/15/2023

• Analyze, estimate and optimize cloud costs using Microsoft Copilot for Azure (preview). Article date. 11/21/2023

• GitHub Copilot announcement in June 2021

• M365 Copilot announcement in March 2023

• D365 Copilot announcement in March 2023

• Windows Copilot announcement in May 2023

• Microsoft Fabric and Copilot in Microsoft Power BI announcement in May 2023

• Announcing a next-generation AI Copilot in Microsoft Power Apps that will transform low-code development. Ryan Cunningham, VP, Power Apps. March 16, 2023. Accessed on 4/27/2023

• Microsoft outlines framework for building AI apps and copilots; expands AI plugin ecosystem. John Roach. Published on May 23, 2023. Accessed on 5/23/2023

• Announcing Copilot Chat. Shannon Monroe. May 1st, 2023. Accessed on 5/23/2023

• Introducing Microsoft Security Copilot: Empowering defenders at the speed of AI. Mar 28, 2023 by Vasu Jakkal - Corporate Vice President, Security, Compliance, Identity, and Management. Accessed on 6/2/2023

• Announcing Microsoft’s AI Customer Commitments. Antony Cook, Corporate Vice President and Deputy General Counsel, Microsoft. Accessed on 6/12/2023

• Microsoft announces new Copilot Copyright Commitment for customers. Brad Smith, Vice Chair and President, Hossein Nowbar, CVP and Chief Legal Officer. Accessed on 9/26/2023

• Microsoft achieves first milestone towards a quantum supercomputer. Chetan Nayak, Technical Fellow and Corporate Vice President of Quantum Hardware. June 21st, 2023. Accessed on 6/21/2023.

• Introducing the Microsoft 365 Copilot Early Access Program and new capabilities in Copilot. By Colette Stallbaumer, General Manager, Microsoft 365 and Future of Work. Accessed on 6/21/2023

• LinkedIn. Pulse. Microsoft Launches New AI Skills Training and Resources as part of Skill for Jobs Initiative. By Kate Behncken Corp. Vice President, Microsoft Philanthropies. Published June 28, 2023. Accessed 6/28/2023

• GitHub Copilot Chat beta now available for every organization. Mario Rodriguez. July 20, 2023. Accessed on 7/20/2023

Links and Resources

• Get Started with GitHub Copilot

• Azure OpenAI Service

• Azure OpenAI Service Pricing

• Azure OpenAI Studio

• GitHub Copilot Licensing

• Copilot for Individuals

• Copilot for Business

• OpenAI Cookbook

• Copilot for CLI

• Embedding Models

• Adopt, extend and build Copilot experiences across the Microsoft Cloud

Language Models

• A Beginner’s Guide to Language Models
• What is a large language model (LLM)?
• A Comprehensive Guide to Build your own Language Model in Python!
• Introduction to Large Language Models
• Language model
• What are large language models?

Tutorials, Quickstarts, and Accelerators

• Quickstart: Chat with Azure OpenAI models using your own data

• Microsoft GitHub. AI Solution Accelerators

• Microsoft Learn AI Skills Challenge

• Azure OpenAI Samples

OpenAI Demos

• OpenAI Codex Live Demo
• OpenAI GPT-4o

Community Demos

• Using Copilot Chat to fix vulnerability discovered by GitHub Advanced Security
• Writing Tests with Copilot
• Copilot : Build a new features in few minutes
• Copilot: code faster with contextual suggestions
• Fun with copilot

Acedemic Lectures

• Harvard. CS50. GPT-4 - How does it work, and how do I build apps with it

GitHub Demos

• GitHub Copilot: your AI pair programmer - GitHub Checkout
• Chat

Microsoft Mechanics

• Microsoft Mechanics. What runs ChatGPT? Inside Microsoft's AI supercomputer. Featuring Mark Russinovich. Accessed on 6/21/2023

Bing DALL-E

• Bing Search Image Creator

Bing Chat

• Bing Chat

Designer with DALL-E

• Stunning designs, made lighting fast with AI

Privacy, Security, Governance, Policy and Legal

• Microsoft Responsible AI

• OpenAI. DALL-E Content Policy. Accessed on 4/27/2023

• OpenAI. Can I sell images I create with DALL-E? Accessed on 4/27/2023

• OpenAI Content Policy. Accessed on 4/27/2023

• Code of conduct for Azure OpenAI Service. Accessed on 4/27/2023

• Microsoft Learn. Transparency note for Copilot data security and privacy in Microsoft Power Platform. Accessed on 6/28/2023

• GitHub Copilot Product Specific Terms. Accessed on 6/28/2023

• GitHub Privacy Statement. Effective date: December 15, 2022. Accessed on 6/28/2023

• GitHub Copilot for Business Privacy Statement. Effective Date: December 7, 2022

• GitHub Copilot Trust Center

Articles

• What is GitHub Copilot? An AI Pair Programmer for Everyone. Daniel Diaz, August 9, 2021. Accessed on 4/17/2023

• GitHub’s Copilot goes beyond code completion, adds a chat mode and more. Frederic Lardinois, March 22, 2023. Accessed on 4/17/2023

• Generative AI set to affect 300 million jobs across major economies. Ars Technica. DELPHINE STRAUSS, FINANCIAL TIMES - 3/28/2023, 8:30 AM. Accessed 4/13/2023.

• Oops: Samsung Employees Leaked Confidential Data to ChatGPT. Gizmodo. Accessed 4/14/2023.

• Research: Quantifying GitHub Copilot's impact on developer productivity and happiness. GitHub Blog. Accessed on 4/14/2023.

• Augmenting In-House Innovation with Startups: Generative AI. ShiSh S. Accessed on 4/14/2023.

• GitHub Copilot X: The AI-powered developer experience. Accessed on 4/17/2023

• Github Copilot for Swift iOS Developers. Rudolf Farkas May 22, 2022. Accessed on 4/19/2023

• Research: How GitHub Copilot helps improve developer productivity. Albert Ziegler July 14, 2022. Accessed on 4/19/2023

• GitHub Copilot for CLI for PowerShell. Scott Hanselman April 25, 2023. Accessed on 4/26/2023

• "The Godfather of A.I." just quit Google and says he regrets his life’s work because it can be hard to stop ‘bad actors from using it for bad things’ by Prarthana Prakash, Fortune Media IP Limited. May 1, 2023 12:55 pm CDT. Accessed on 5/2/2023

• DeepMind cofounder warns governments seriously need to find solutions for people who lose their jobs to A.I. CHRISTIAAN HETZNER May 10, 2023. Fortune Media IP Limited. Accessed on 5/21/2023.

• GPT-4 Will Have 100 Trillion Parameters — 500x the Size of GPT-3. Are there any limits to large neural networks? By Alberto Romero. Published in Towards Data Science. Sep 11, 2021. Accessed on 6/2/2023

• Microsoft has over a million paying Github Copilot users: CEO Nadella. Tiernan Ray, Senior Contributing Writer Oct. 25, 2023. Accessed on 10/27/2023

Articles on AI and Compute

• Microsoft signs deal for A.I. computing power with Nvidia-backed CoreWeave that could be worth billions. Jun 1, 2023. CNBC. by Jordan Novet. Accessed on 6/2/2023

• Nvidia's founders came up their trillion-dollar idea while spending hours drinking Denny's coffee and eating Grand Slam breakfasts: 'We were not good customers'. By Grace Mayer. MSN. Insider.com Accessed on 6/2/2023

What Research Firms are saying

• The state of AI in 2022—and a half decade in review. December 6, 2022 | Survey. Accessed on 6/2/2023

What the C-Suite is saying

• "Satya Nadella: Microsoft's Products Will Soon Access Open AI Tools Like ChatGPT | WSJ". Wall Street Journal. Accessed on 4/14/2023.

• "OpenAI CEO, CTO on risks and how AI will reshape society". ABC News. Accessed on 4/14/2023.

• "OpenAI CEO Sam Altman | AI for the Next Era". Greylock. Accessed on 4/14/2023.

• "Fmr. Google CEO Eric Schmidt on the Consequences of an A.I. Revolution". Amanpour and Company. Accessed on 4/14/2023.

• "Bill Gates on AI and the rapidly evolving future of computing". Microsoft. Accessed on 4/14/2023.

• "OpenAI GPT-4: The Most Advanced AI Yet - Live with Tesla & Elon Musk". Tesla. Accessed on 4/14/2023.

• "OpenAI Co-founder Greg Brockman on ChatGPT, DALL·E and the Impact of Generative AI | SXSW 2023". SXSW. Accessed on 4/14/2023.

• Sam Altman: OpenAI CEO on GPT-4, ChatGPT, and the Future of AI | Lex Fridman Podcast #367

Certifications and Learning Paths

• Microsoft Azure AI Fundamentals: Get started with artificial intelligence

• Developing plugins with Semantic Kernel and Azure OpenAI

Introduction to large language models

Hello World Copilot Examples Usage

A few code files will include how to run comments inline.

Compiling a .cs file

."C:\Program Files\Microsoft Visual Studio\2022\Enterprise\MSBuild\Current\Bin\Roslyn\csc.exe" .\main.cs

Running an executable

.\main.exe  

Lessons Learned

1. The Coldfusion language may not supported by Copilot. It did take me some time to get it to work with it. It did finally do write me some code.
2. I'm having a bit of a challenge to write VB. VBScript was straight forward.
3. Copilot not much of a help yet when it comes to .bicep files.

Contributing

This project welcomes contributions and suggestions. Most contributions require you to agree to a Contributor License Agreement (CLA) declaring that you have the right to, and actually do, grant us the rights to use your contribution. For details, visit https://cla.microsoft.com.

When you submit a pull request, a CLA-bot will automatically determine whether you need to provide a CLA and decorate the PR appropriately (e.g., label, comment). Simply follow the instructions provided by the bot. You will only need to do this once across all repos using our CLA.

This project has adopted the Microsoft Open Source Code of Conduct. For more information see the Code of Conduct FAQ or contact opencode@microsoft.com with any additional questions or comments.

Legal Notices

Microsoft and any contributors grant you a license to the Microsoft documentation and other content in this repository under the Creative Commons Attribution 4.0 International Public License, see the LICENSE file, and grant you a license to any code in the repository under the MIT License, see the LICENSE-CODE file.

Microsoft, Windows, Microsoft Azure and/or other Microsoft products and services referenced in the documentation may be either trademarks or registered trademarks of Microsoft in the United States and/or other countries. The licenses for this project do not grant you rights to use any Microsoft names, logos, or trademarks. Microsoft's general trademark guidelines can be found at http://go.microsoft.com/fwlink/?LinkID=254653.

Privacy information can be found at https://privacy.microsoft.com/en-us/

Microsoft and any contributors reserve all other rights, whether under their respective copyrights, patents, or trademarks, whether by implication, estoppel or otherwise.