Год: 2025

Do you want to learn data wrangling with Python on a budget? No worries, there are (at least) five free courses that’ll provide you with solid knowledge.

Microsoft CEO Satya Nadella revealed at the Microsoft AI Tour in Bengaluru that OpenAI’s ChatGPT is the largest user of Cosmos DB, Microsoft’s database service, designed to build scalable and AI-ready apps.

“If you look at ChatGPT, they are some of the biggest users of Cosmos because [it] serves as a stateful application,” Nadella said. This underscores Cosmos DB’s role in managing extensive real-time interactions and session data required by ChatGPT, a chatbot that handles millions of simultaneous conversations.

Nadella emphasised cloud-native databases such as Cosmos DB, SQL Hyperscale, and Fabric for modern AI workloads. He said that AI doesn’t sit on its own and, instead, relies on a comprehensive compute stack to function effectively.

Cosmos DB’s scalable architecture supports high-throughput, low-latency demands during both AI training and inference, thereby ensuring responsive and stateful interactions for applications like ChatGPT.

Linking AI’s progress to Moore’s Law, Nadella noted that scaling laws remain strong for both training and inference and highlighted the significance of test-time compute”, where databases like Cosmos DB must handle high-throughput and low-latency demands post-training.

He pointed out that merging operational and analytical data in the cloud enables AI-driven apps to access high-volume data reliably and in real time.

Nadella attributed ChatGPT’s success to the foundational role of databases in AI and said that Cosmos DB integrates operational and analytical data to provide reliable, high-volume access in real time.

According to the company, this focus on robust data infrastructure aligns with its broader strategy, including a $3 billion investment in Azure data centres across India, where the aim is to expand global capacity to meet rising AI demands.

These investments reinforce Cosmos DB’s position as a backbone for next-generation AI applications.

At last year’s Microsoft Build, the company introduced Microsoft Fabric, real-time intelligence to its AI analytics platform. This update integrated SQL Server into Fabric databases and merged operational and analytical data within a single unified platform.Moreover, the data API builder for Azure Cosmos DB was released to the public. This open-source tool requires no coding to set up secure GraphQL endpoints.

The post ‘ChatGPT is Cosmos DB’s Largest User’ appeared first on Analytics India Magazine.

AMD has unveiled its highly anticipated next-generation GPUs, including the Radeon RX 9070 line for gaming and content creation, on Jan. 6 ahead of CES 2025 in Las Vegas. Powered by the cutting-edge RDNA 4 architecture, the new Radeon lineup promises significant advancements in AI computing, media encoding, and ray tracing capabilities. These GPUs are expected to hit the market in Q1 2025.

AMD’s GPUs compete with NVIDIA’s, which are also expected to see a refresh at CES.

“As consumers and professionals increasingly recognize the productivity benefits of AI PCs, AMD is further increasing its performance leadership in the market,” Jack Huynh, senior vice president and general manager of AMD’s computing and graphics group, said in a press release. “With the next generation of AI-enabled processors, we are proliferating AI to devices everywhere, and bringing the power of a workstation to thin and light laptops.”

The Radeon RX 9070 series is built on RDNA 4 architecture

The Radeon 9000 series GPUs are built to complement AI workloads.

The first two cards in the series, Radeon 9070 XT and Ryzen 9070, will be available in hardware from Acer, Asus, and other manufacturers. Details about the new series remain limited, but the RDNA architecture — built on the 4nm manufacturing process — promises significant advancements. The GPUs are expected to excel in AI computing, offer improved media encoding capabilities, and deliver improved ray tracing for gaming and graphics creation.

Not much information is available about the Radeon RX 9070 series yet. However, AMD said over 100 enterprise platform brands will use AMD Pro technology through 2025.

The company also announced the Ryzen 9000 HX series, including the beefy Ryzen 9950X3D for gaming and content creation.

SEE: Shortly before CES 2025, OpenAI announced its new phase: the quest for “superintelligence.”

Ryzen AI Max and others bring AI performance to laptops

Overall, Ryzen continues to devote significant attention to AI PC products, including the third-generation AMD Ryzen AI and Ryzen AI 300 Pro series processors announced at CES.

These chips compete with Qualcomm’s and Intel’s Copilot+ PC processors. AMD claims that the Ryzen AI 7 350 and Ryzen AI 5 340 perform at 50 TOPS at peak performance. Options range from the 6-core, 12-thread AI Max Pro 380 to the 16-core, 32-thread AI Max/AI Max Pro 395 build. The AI Max and AI Max Pro series offer up to 256 GB/s bandwidth. HP and Asus are on board to sell laptops and mini workstations equipped with these chips.

AMD announced the AI Max and AI Max Pro series processors for gaming, scheduled for the first half of 2025. Although mobile gaming may be the primary use case, these heavy-duty graphics engines may find their way into compact workstations for graphic designers, game developers, and other graphics-intensive professional use cases.

Ryzen Z2 and Ryzen 200 processors coming soon

Other announcements from AMD at CES 2025 included:

• The Ryzen Z2 Series for handheld PC gaming is available in the first quarter of 2025.
• The Ryzen 200 Series processors for IT environments and other mainstream commercial uses are available in the second quarter of 2025.

“AMD Ryzen CPUs offer the best TCO [total cost of ownership], delivering exceptional performance and efficiency,” said Robert Kochheim, global portfolio manager of digital workplace at Shell in an AMD press kit.

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.

The post Microsoft Shows How to Code in Kannada appeared first on Analytics India Magazine.

Understanding the latest project to build speech-to-speech with open source technologies.

The Uttar Pradesh government, in collaboration with Google Cloud India, has launched the Gemini-powered, Beckn-enabled Uttar Pradesh Open Network for Agriculture. The initiative aims to support millions of farmers by offering a single platform for advisory services, credit, mechanisation, and market linkages.

This first-of-its-kind digital public infrastructure (DPI) for agriculture leverages Google’s DPI-in-a-box solution and the Beckn protocol, enabling interoperability and decentralised discovery of services. Farmers can access the platform using voice commands in languages like Hindi, Bengali, Telugu, Kannada, Gujarati, and Punjabi, with more languages to be added in 2025.

“This agreement demonstrates the UP government’s commitment to good governance and digital empowerment,” said Manoj Kumar Singh, state chief secretary. “The Open Network for Agriculture is a major step toward doubling farmers’ income. By leveraging AI, it addresses the digital divide and equips our farmers with the tools and resources they need to thrive in today’s day and age.”

Unlike traditional marketplaces, this open and decentralised platform allows service providers, from input suppliers to financial institutions, to connect with farmers directly. It also enables anyone to curate these services, build farmer-facing solutions and foster innovation.

“We are proud to partner with the government of UP on this pioneering initiative,” said Bikram Singh Bedi, vice president and country MD of Google Cloud India. “Technology is critical for accelerating agricultural progress, and this Gemini-powered Open Network offers an open, decentralised approach that innovators can leverage to build solutions for farmers.”

Sujith Nair, CEO of FIDE (Foundation for Interoperability in Digital Economy), which developed the Beckn protocol, said, “Beckn aims to unlock access and enable value exchange at population scale. We’re thrilled to see the UP government and Google Cloud India leverage Beckn protocol to launch a state-wide Open Network for Agriculture in India.”

This initiative underscores the state’s efforts to digitise its agricultural ecosystem and bridge the digital divide, thereby making it a model for scalable open solutions in India.

The post Google Collaborates with UP Government to Launch Gemini-Powered Open Network for Farmers appeared first on Analytics India Magazine.

Using AI for coding is a double-edged sword. While tools like Cursor or GitHub Copilot and chatbots like Claude or ChatGPT can help write code faster, they also come with hallucinations and errors, making the use of generated code in production risky.

Junior developers, hired amid this coding revolution, often fail to grasp the issue or ignore it in a bid to complete tasks faster. A Reddit thread recently highlighted concerns from senior engineers who see both the potential and pitfalls of AI reliance in development. It is almost like copying and pasting code without understanding the meaning behind the code.

A senior engineer shared his experience with two junior developers on Reddit. While one used AI responsibly—like a ‘search engine that doesn’t get mad at you’—the other seemed to rely entirely on AI for generating code, often copying and pasting without reviewing or understanding it. The results included ‘robotic, but technically correct code’ and occasional nonsensical snippets like a for-loop doing absolutely nothing.

This over-reliance on AI led to recurring issues in code reviews. As the engineer noted, “I wouldn’t have issues if I didn’t make repeat comments like, ‘I’m not sure what you’re trying to do here?’ and then getting a response like, ‘idk, I’ll remove this.’”

Copy, Paste, Move-on?

Before tools like ChatGPT became developers’ trusty assistants, they would often head to Stack Overflow with their doubts and copy-paste the responses. The now-senior developers, who were getting started with coding back then, are the ones guilty of this crime.

Speaking with AIM, Adarsh Shirawalmath, CEO of Tensoic and an experienced programmer, said that overreliance on such tools is not good for developers. “I don’t think they overuse the tools, but they use them often without understanding what the output code is,” Shirawalmath said.

He added that the practice was rampant and could become a problem for the companies these people worked at. “Sometimes ChatGPT’s responses are hallucinations, and the model is so confident in its answer; it’s tough to debug when something goes wrong,” he added.

However, companies cannot ban the use of AI since they want to be part of the revolution. Sergey Brin, the co-founder of Google, earlier said that he believes developers are not using AI coding tools enough. He also said that writing code from scratch feels difficult after relying on AI coding tools. Hence, finding the right balance is the core issue for modern developers.

Siddharth Sharma, former CTO of Shaadi.com, cancelled his Cursor subscription as it was negatively affecting his thought process. “In general, I found the LLM never pointed to a refactoring that would make the code less verbose. And this was painful as it happened. It is going to be 10 times more painful in teams and large codebases,” he warned.

This has led to many companies being critical of their employees using AI tools. A major reason behind Indian companies’ cautious approach to using AI coding tools is privacy. Many fear that these tools, especially those developed by large corporations, might expose sensitive or proprietary code to third parties.

What Can Senior Developers Do?

The Reddit thread had reactions pouring in from developers who shared similar frustrations. One commenter emphasised the importance of intentionality in coding: “You have to know why it’s the right answer; otherwise, you’re deluding yourself.”

Many advised senior developers to address such behaviours in private 1:1 discussions to stress the need for understanding the “why” behind the code. “People attempting to merge code into main that they may be ‘kinda sorta’ half understand why it works is legitimately frightening,” a developer added.

Adithya S Kolavi, founder of Cognitive Labs, told AIM that he had seen a lot of junior developers using AI tools, and there is always a risk of overreliance on such tools.

“When I explore a new technology, I prefer taking the traditional route—reading documentation, exploring forums, and deeply understanding the concepts. I do use AI as a learning tool, like a virtual mentor, to clarify doubts or explain concepts, but not to generate entire blocks of code,” said Kolavi.

He added that blindly relying on AI to write code without understanding the underlying technology can be detrimental. “I’ve seen situations where developers get stuck in a frustrating loop—asking AI to generate code, receiving incorrect or misleading output, and then repeatedly prompting it to fix itself, only to waste time running in circles,” Kolavi added.

Several developers, fresh out of college and learning to code with AI, are also unable to land jobs due to this. While it is fine to get the help of AI coding tools to learn how to code, overreliance is where the issue begins. “Wake me up when companies start hiring these clueless modern ‘developers’,” said Santiago Valdarrama, founder of Tideily and an ML teacher.

Andrej Karpathy, founder of Eureka Labs, had once accepted that coding with AI led him to introduce new bugs. “It’s slightly too convenient to just have it do things and move on when it seems to work,” he said.

Meanwhile, it has been reiterated that using AI for coding would turn developers into 10x developers. Many seasoned developers agreed that using AI isn’t inherently wrong—it can speed up development and help solve complex problems. However, juniors must take the responsibility for reviewing and understanding the output. “I go line by line through each code snippet, understand it, and adjust where needed,” advised a developer.

Senior engineers in the thread warned that juniors relying excessively on AI are not only stunting their own growth but also creating long-term issues for their teams. One contributor shared a hard decision they faced. “I had a junior doing this. They kept doing it after repeated warnings. Firing anyone sucks, but I was going to lose my best people if I kept him on.”

The post Junior Devs are Letting AI Do the Heavy Lifting—But at What Cost? appeared first on Analytics India Magazine.