Data Science News, Latest Updates and Recent Announcements on Techminis

A naukri.com initiative

New

Home

Data Science News

VentureBeat

101

Image Credit: VentureBeat

The code whisperer: How Anthropic’s Claude is changing the game for software developers

AI assistants have become indispensable tools in the global software development market, with one of the products leading this shift being the AI model, Anthropic’s Claude.
Claude's coding-related revenue has surged 1,000% over the past three months, with software development accounting for over 10% of all Claude's interactions.
Claude's ability to analyze up to 200,000 tokens of context, equivalent to about 150,000 words, has helped it generate a lot of value compared to previous models and maintain coherence throughout an entire development session.
OpenAI, Google's Gemini, and Meta's Llama 3.1 have doubled down on developer tools to compete with Claude.
Claude has led to significant productivity gains with GitLab reporting 25-50% efficiency improvements among its development teams and Sourcegraph seeing a 75% increase in code insertion rates after switching to Claude as its primary AI model.
Claude is changing who can write software as it has become more accessible to non-programmers, allowing every department to solve its own problems.
AI-coded code assistants will fundamentally change the way software is created, with Gartner predicting that by 2028, 75% of enterprise software engineers will use AI code assistants.
As AI handles routine tasks, developers are freed to focus on system architecture, code quality, and innovation.
The impact of AI on coding assistants extends far beyond individual developers, with major tech companies reporting significant benefits such as Amazon using its AI-powered software development assistant to migrate over 30,000 production applications.
However, the effects of AI coding assistants are not uniformly positive, with studies suggesting that while AI can accelerate certain development tasks, it may also introduce new challenges in code quality and maintenance.

Read Full Article

6 Likes

Medium

188

Image Credit: Medium

AI Engineers Future

Companies are adopting AI-driven solutions to keep up with current trends.
AI chatbots are being implemented on company websites for internal and client interaction.
The demand for AI engineers who possess skills in Data Science, AWS, and Python is high.
The future of the IT industry is focused on AI, making AI engineers essential.

Read Full Article

11 Likes

Medium

Image Credit: Medium

A Merry Medium Christmas

The holiday season is approaching, and emotions are varied amongst people.
While some are excited, others are indifferent or disinterested.
Many people are planning to spend time with family, highlighting the significance of familial bonds.
There are concerns about performance and not meeting certain standards, causing some to question their effectiveness.

Read Full Article

1 Like

Analyticsindiamag

128

Image Credit: Analyticsindiamag

Infosys Partners With Indian National Academy of Engineering to Transform Engineering Education in India

Infosys Foundation signed a four-year MoU with INAE to enhance engineering education in India.
The collaboration aims to upskill educators and provide students with internships and mentorship opportunities.
Educators will receive in-person and virtual training to improve technical expertise and student engagement.
INAE will offer project-based internships and mentorships to students pursuing B.Tech and BE courses.

Read Full Article

7 Likes

Discover more

Unite

223

Image Credit: Unite

How AI is Making Sign Language Recognition More Precise Than Ever

A team at Florida Atlantic University has used artificial intelligence and a new approach to improve Sign Language recognition's accuracy. Their technology can now recognize American Sign Language (ASL) alphabet gestures with precision. The researchers have carefully integrated transfer learning, meticulous data creation, and precise tuning for the system's remarkable performance. MediaPipe and YOLOv8, when put together, provide a 98% accuracy rate and 99% F1 score. The next step is to teach this system wider-ranging hand shapes and gestures and ensure it works in any environment while being fast for practical and real-time uses.
Previous sign language recognition technology has struggled with Accuracy because of the complex languages and unique grammar and syntax of each sign language worldwide. Understanding each element of sign language, which includes facial expressions and body language, is more challenging than just hand movements. The Florida Atlantic University's College of Engineerging and Computer Science focused on recognizing ASL alphabet gestures with clarity using 21 keys precisely marked on the hands in the dataset of 29,820 static images.
MediaPipe is an expert hand watcher capable of tracking finger movements and hand positions accurately. This ability to provide accurate hand landmark tracking makes it an exceptional option for sign language interpreters to detect hand gestures' precise coordinates. The YOLOv8 is an advanced object detection technology that understands what these gestures represent after interpreting the tracked points of each sign and predicting the probability of a hand gesture being present within each image.
The system understands each person's hand gestures, detecting and classifying American Sign Language gestures with 98% precision. This technology works flawlessly in different and elaborate lighting, different hand positions, and different people's signs, making it quite versatile and futuristic. The researchers are working to cover all the aspects previously missed and teach the system even more hand shapes and gestures, making it smoother, quicker, and reliable for all types of people just like their real hands.
The success of this project will open up exciting possibilities to enhance communication and understanding among people, regardless of any language barriers. This technology can be used to create tools that would improve communication for the deaf and hard-of-hearing community. The researchers are hoping to work on the speed and reliability of communication between hearing-impaired and normal people and how to integrate this technology seamlessly into everyday life.
The translation of sign languages can be highly complex because each sign represents a unique concept. Different grammars and syntaxes unique to each language worldwide make building accurate sign language translations difficult. Being able to break these communication barriers through this technology is essential for people to connect on different aspects of life.
Such progress in communication technology represents a step closer to real-world applications in either healthcare or education, where instant language translation could make life easier for both deaf and hearing communities. The cumulative efforts of engineers and Computer Scientists to improve communication demonstrate the versatility of programming and AI in technology's continued growth and development.

Read Full Article

13 Likes

Analyticsindiamag

4.9k

Image Credit: Analyticsindiamag

Ola Krutrim is Now Offering Free Credits to all IIT Students and Faculty

Bhavish Aggarwal’s Ola Krutrim is now offering free credits for Krutrim for students and faculties of all IITs in the country.
Ola Krutrim has announced free कृत्रिम (Krurtrim) credits for IIT students and faculty.
Ola had launched Krutrim AI Cloud, its own cloud platform for enterprises, researchers, and developers.
Ola Consumer, led by Aggarwal, is testing 10-minute food deliveries in Bengaluru using the ONDC platform.

Read Full Article

21 Likes

Analyticsindiamag

231

Image Credit: Analyticsindiamag

Google Deepmind’s New Benchmark Evaluates Factuality of LLMs

A new benchmark tool, FACTS Grounding, was announced by Google DeepMind and Google Research.
FACTS Grounding evaluates the factual accuracy of LLMs (language models).
It aims to address challenges such as misinformation and hallucination in AI-generated content.
The benchmark involves testing models on over 1700 tasks and includes a leaderboard on Kaggle.

Read Full Article

13 Likes

Analyticsindiamag

304

Image Credit: Analyticsindiamag

Is It Too Soon to Talk About AGI and Regulation?

AGI (artificial general intelligence) is a leap forward in mimicking the cognitive abilities of the human brain and bridging the gap between human and machine intelligence. However, the means to AGI and its regulation remain a topic of debate amongst experts.
Many AI systems have evolved from deep learning to LLMs (large language models), enabling accessible and democratic AI technology. However, the role of LLMs as a pathway to AGI has been a controversial topic.
AGI is designed to execute tasks autonomously, enabling systems to learn effectively and behave like humans. This evokes the question of singularity where machine intelligence can surpass human intelligence.
AGI's practical applications in healthcare offer unique solutions for disease management, but the development of AGI must go hand-in-hand with governance models to ensure ethical deployment and alignment with human values.
India as a hub of AI innovation needs to adopt a flexible regulatory framework that allows innovation to thrive while upholding foundational principles of human-centricity, accountability, and fairness.
Experts believe a proactive governance model, where startups and enterprises incorporate ethical practices by default, would be more effective in enabling innovation while ensuring compliance with ethical standards.
Regulators, innovators, and all stakeholders' collaboration is essential for establishing trust and preemptively addressing potential pitfalls, ensuring AGI serves humanity positively.
Time will tell if AGI is achievable, but current progress in the field proves that responsible AI deployment and regulation should be key objectives in creating a future where innovation and responsibility coexist.
Experts advise that everyone involved in regulating AI technology should deploy an ML model, sit with someone, or at least know how to deploy it.

Read Full Article

18 Likes

Analyticsindiamag

325

Image Credit: Analyticsindiamag

Meet Shivaay, the Indian AI Model Built on Yann LeCun’s Vision of AI

Indian engineering students Rudransh Agnihotri and Manasvi Kapoor have launched an AI startup called FuturixAI.
They developed Shivaay, an AI model with 4 billion parameters built on a joint embedding architecture.
Shivaay outperforms larger state-of-the-art models and excels in reasoning and mathematical calculations.
The goal of FuturixAI is to empower Indian developers and businesses to build their own AI agents and applications.

Read Full Article

16 Likes

Analyticsindiamag

281

Image Credit: Analyticsindiamag

AI Agents to Upgrade Chips in India

Synaptics and embedUR Systems have committed to the development of energy-efficient and compute-efficient edge devices in India.
Through its software solutions, the companies could bring the cost of AI-enabled microcontrollers down to a range of $10 to $20.
The current solutions manufactured by companies such as NVIDIA are priced around $10,000 per chip.
Indian employees are essential as Synaptics moves to expand its workforce: about 25% of the tech company's staff work from Bengaluru.
EmbedUR Systems operates in Chennai, mastering the process required for system improvements and low attrition rates.
The companies hope that their collaboration could help Indian machinists produce energy-efficient and computing-efficient edge devices,
Cost-efficient, AI-enabled appliances could positively impact Indian society. For instance, LG's ThinQ washing machines, which retail at around $3,000, often entailing high-end consumers, could be reimagined with affordably available products.
Partnering with embedUR Systems, Synaptics plans on using AI models designed to reduce power and resource requirements.
According to Synaptics CEO Michael Hurlston, the distinction between cloud-based processing and device-based processing will blur as the tech industry shifts towards efficient, localised edge AI solutions.
India has the potential to lead in this emerging domain of edge AI by capitalising on local talent and adopting globally competitive solutions.

Read Full Article

16 Likes

Analyticsindiamag

178

Image Credit: Analyticsindiamag

Synaptics, EmbedUR Collaborate to Build Cost-Effective AI Chips For India

Synaptics and embedUR are collaborating to create energy-efficient and compute-efficient edge devices for India's semiconductor industry.
The two companies aim to manufacture low power, low-cost neuro-processing units and microcontrollers with embedded software that can operate AI.
The expectation, according to Synaptics' CEO, Michael Hurlston, is to replace traditional MCUs completely with chips from embedUR to provide the same product more affordably, supported by AI capabilities.
The cost of AI-enabled MCU is estimated to fall between $10 and $20 using Synaptics chipsets and embedUR's efficient software solutions.
Companies like NVIDIA offer high-performance semiconductors, such as the Blackwell processors, which can cost up to $10k per chip.
By optimising software, and reducing the overall bill of materials, AI could easily be added to existing mid-level and low-cost devices such as washing machines without significantly driving up prices.
Synaptics already has 25% of its global workforce stationed in Bengaluru intending to double that number over the next three years to support the growth of its engineering skills headquartered there.
embedUR's proprietary AI model framework, ModelNova, strengthens the company's expertise in connectivity and embedded software, by providing lightweight, efficient models designed to run on low cost, resource-constrained devices.
The Synaptics-embedUR partnership is designed to democratise the application of AI, with potential implications for everyday devices in the mass market, particularly in India.
Edge AI aims to remove the need for large-scale data centres by running machine learning (ML) algorithms using software and models devices at the device level, potentially leading to significant advancements for India's semiconductor industry.

Read Full Article

10 Likes

Analyticsindiamag

221

Image Credit: Analyticsindiamag

Meet Shivaay, the Indian AI Model Built on Yann LeCun’s Vision of AI

Indian engineering students Rudransh Agnihotri and Manasvi Kapoor have developed an AI model called Shivaay.
Shivaay is built on a joint embedding architecture and leverages three models for data.
The model outperforms larger state-of-the-art models on benchmarks and is excellent at reasoning and mathematical calculations.
The goal of the team is to empower Indian developers and businesses to build their own AI agents and applications without relying on foreign models like GPT.

Read Full Article

13 Likes

Analyticsindiamag

Image Credit: Analyticsindiamag

Trump Appoints Indian-American Sriram Krishnan as Senior Policy Advisor for AI

US President-elect Donald Trump has appointed Sriram Krishnan, an Indian-American former venture capitalist, as the senior policy advisor for artificial intelligence at the White House Office of Science and Technology Policy.
Krishnan will help shape AI policy across government and work with the President's council of advisors on science and technology.
He has prior experience in consumer AI and crypto investments and has worked with companies like Twitter, Meta, and Snap.
Krishnan, along with David Sacks, will work closely with Elon Musk in the Trump administration for AI and government efficiency.

Read Full Article

21 Likes

Medium

Image Credit: Medium

Python and Probability Mathematics

Probability is a foundational aspect of mathematics that helps analyze uncertainties.
Python makes exploring probability concepts straightforward with its versatile tools and libraries.
Some real-world applications of probability in Python include predicting customer behavior, quality assurance, machine learning, genetics, network reliability, scheduling tasks, lottery odds, financial modeling, and predicting customer purchases or election results.
Python's probability capabilities are valuable in various fields and offer versatile applications for analyzing uncertainties.

Read Full Article

1 Like

For uninterrupted reading, download the app