At the Microsoft AI Tour in Bengaluru, the company took the stage to demonstrate GitHub Copilot Workspace and its capabilities to assist users in writing code through natural-language queries.
The company did not miss a chance to add a local flavour to the event. Karan MV, director of international developer relations at GitHub, demonstrated that Copilot’s Workspace can understand Indian languages, including Hindi and Kannada, while writing code.
The feature was introduced in May last year. Microsoft CEO Satya Nadella said, “Think about this – every person can now start programming, whether it’s in Hindi, Brazilian, or Portuguese, and bring back the joy of coding in their native language.”
GitHub Copilot Workspace is an environment that assists developers with AI through their coding workflow.
During the demonstration, Karan wanted to add functionality that allows images to be added to a product website. He asked Copilot in Hindi, “Ek page pe image upload karne ke liye koi idea batiye,” which translates to “Any idea for uploading images on a page?”
Copilot Workspace then generated ideas, such as adding a file input element, using drag-and-drop, or creating a third-party library area inside the website. After selecting the preferred idea, Copilot Workspace generated a list of tasks needed to add the feature and then generated the code for it.
Karan then went ahead and asked to add a preview of the uploaded image. He asked in Kannada, “Image upload madidamele adara preview thorsi,” which translates to “Show me a preview of the image once it is uploaded”. Copilot Workspace successfully added code that would display a preview of the image on the website.
Kannada is the official and most spoken language of Karnataka. The capital city of Bengaluru is home to over 360,000 software engineers.
GitHub said that 17 million Indian developers use the platform, second only to the United States of America. However, in 2021, it was reported that a whopping 67% of engineers graduating from Indian colleges do not possess the necessary English-speaking skills.
Good prompts require a decent mastery of the language being used to prompt, and the best results are directly proportional to the clarity of the input commands. “Remember when we used to make fun of ‘primitive’ people who believed that using the right kinds of words will result in magic and incredible creations?” said Bojan Tunguz, a former AI engineer at NVIDIA, indicating the importance of the right choice of words.
Therefore, if an AI tool wants to assist the wealth of developer talent in India, it is only going to pose a second challenge if English is the only supported language.
Over the past few months, GitHub has released over 100 new additions to the Copilot Workspace, repeatedly suggesting that it is the ‘developer environment of the future’. The newly announced build and error repair feature suggests potential solutions to code errors and provides an option to fix the code manually or let Copilot Workspace rectify it automatically.
GitHub also introduced enhancements to the Brainstorming Mode in Copilot Workspace. It lets users collaborate and explore all the repositories, suggest solutions for issues, ask questions and help improve the overall problem-solving process.
Last month, GitHub announced that it is making Copilot available for free within Visual Studio Code (VS Code). “We’re seeing a surge in Copilot Free activation in India and across APAC. It’s incredible to see (sic),” CEO Thomas Dohmke said in a post on X.
Note that GitHub Copilot is an AI coding assistant built inside editors, whereas GitHub Copilot Workspace is a broader platform that integrates AI throughout the entirety of the development workflow.
Meanwhile, Microsoft just announced its largest investment in India yet, a $3 billion commitment to expand Azure’s infrastructure in the country. The company also revealed that it will train 10 million people in AI by 2030 as a part of its ADVANTA(I)GE INDIA initiative.
