A naukri.com initiative
Data Science News
Eweek
1w
337
Image Credit: Eweek
Top 10 AI Software Tools of 2025: Unleashing Business Innovation
- The top AI software applications offer distinct capabilities from increasing productivity to elevating customer experiences. Here are our picks for the top AI software for 2025 including Grammarly, Tidio, ChatGPT, Fireflies.ai, Microsoft Copilot, Synthesia, Salesforce Einstein, Claude, Jasper, and IBM watsonx.
- Grammarly, a writing assistant, offers suggestions to tailor content, tone, length, and formality to match brand. It detects AI in content as part of its AI features. Monthly access costs $12 per user.
- Tidio's AI-powered customer communication platform offers live chat and chatbot functionalities allowing businesses to automate interactions and improve response times on multiple channels. Monthly subscription starts at $29 for 100 handled conversations.
- ChatGPT engages in human-like conversations, assists in coding, generates images, and answers questions. Monthly pricing starts at $20 per user for its basic plan.
- Fireflies.ai allows automatic summarization and analysis of voice and video conversations. The platform has advanced Conversation Intelligence capabilities that include speaker talk time tracking, sentiment analysis, and customizable topic trackers. Paid plans start at $18 per seat, per month.
- Microsoft Copilot generates content and automates tasks within Microsoft environments. The platform generates images through its Designer integration. Its monthly subscription starts at $20 per user.
- Synthesia produces professional-quality videos with AI-generated avatars and voiceovers. The platform has a one-click translation feature and monthly subscription costs $29 per user.
- Salesforce Einstein is an advanced AI tool integrated into the Salesforce platform that can automate tasks and generate insights using predictive analytics. The subscription costs $500 per user, per month.
- Claude is an AI tool with a focus on ethical outcomes as much on high performance. Its emphasis on ethical AI development makes it reliable for priority in AI use. Monthly access costs $20 per user.
- Jasper, an AI tool for digital marketing teams, enables marketing teams to quickly respond to trends and audiences and keep marketing materials engaging. Its monthly plans start at $49 per seat, per month.
- IBM watsonx is an enterprise AI and data platform for extracting valuable insights from large datasets. Monthly subscription to this tool starts at $1,050.
Read Full Article
20 Likes
Analyticsindiamag
1w
414
Image Credit: Analyticsindiamag
OpenAI Unveils ChatGPT Projects, Drops AGI Easter Eggs
- OpenAI unveiled a new feature for ChatGPT called Projects.
- Projects allows users to upload files, set custom instructions, and tailor ChatGPT to suit specific projects.
- The feature was rolled out instantly during the live stream and complements ChatGPT's conversation search.
- OpenAI teased an Easter egg related to AGI in the demonstration.
Read Full Article
24 Likes
Analyticsindiamag
1w
247
Image Credit: Analyticsindiamag
After Gemini 2.0, Google Cooks with NotebookLM
- After the launch of Google Gemini 2.0, Google is introducing new features to NotebookLM.
- NotebookLM 2.0 includes a redesigned interface, interactive audio features, and a premium subscription called NotebookLM Plus.
- Users can now interact directly with audio hosts through the Interactive Audio Overview feature.
- The premium plan, NotebookLM Plus, offers higher usage limits, custom responses, team notebooks, and better security.
Read Full Article
14 Likes
Medium
1w
98
Image Credit: Medium
How To Give Positive And Constructive Feedback As An Introverted Engineering Manager
- Giving constructive feedback is challenging for introverts and for engineering managers. It is a critical component of your job to help move the team forward.
- Feedback conversations that only include constructive feedback or only positive feedback are problematic.
- It is advisable to give both kind of feedback to provide a balanced approach. This also helps the receiver see that performance is a work in progress.
- Preparing for feedback meetings beforehand makes it easier for introverted engineering managers to remain calm and focused.
- Include small talk at the beginning of the feedback meeting to make everyone feel more comfortable.
- When giving feedback, focus on the issue, not the person.
- Shift the conversation to solve the problem as a team challenge rather than singling out an individual for something they did wrong.
- Developing an environment in which feedback is a regular part of the process can help to prevent defensiveness and promote growth.
- Introverted engineering managers should not try to behave like extroverted "bosses."
- Feedback is a two-way street, and managers should solicit insights from their team about how they can provide better support.
Read Full Article
5 Likes
VentureBeat
1w
328
Image Credit: VentureBeat
Shutterstock pioneers ‘research license’ model with Lightricks, lowering barriers to AI training data
- Shutterstock and Lightricks have partnered to pioneer a new 'research license' model for accessing AI training data.
- The research license allows companies to test and experiment with smaller licenses before committing to more expensive commercial licenses.
- This approach aims to lower barriers to entry for AI companies and promotes ethical AI development.
- The partnership also addresses concerns about copyright infringement and ensures fair compensation for content creators.
Read Full Article
19 Likes
Analyticsindiamag
1w
256
Image Credit: Analyticsindiamag
Cohere Launches Command R7B to Disrupt Enterprise AI Market
- Canadian AI startup, Cohere, has launched Command R7B, the smallest model in its R series of large language models (LLMs), targeting businesses with a focus on speed, cost efficiency, and flexibility.
- Command R7B is suitable for deployment on low-end GPUs, CPUs, and MacBooks, offering features such as retrieval-augmented generation (RAG) with inline citations, multilingual capabilities, and performance across math, code, and reasoning tasks.
- Cohere's Command R7B is accessible via the Cohere Platform and HuggingFace, priced at $0.0375 per million input tokens.
- Cohere recently secured a $240 million investment to build a multibillion-dollar AI data centre in Canada and specializes in developing LLMs for enterprise applications.
Read Full Article
15 Likes
Eweek
1w
209
Image Credit: Eweek
Large Language Model: A Guide To The Question ‘What Is An LLM”
- Large language models (LLMs) are artificial intelligence systems trained on vast amounts of data that can understand and generate human language.
- LLMs use deep learning technology and natural language processing (NLP) to perform an array of tasks, including text classification, sentiment analysis, code creation, and query response.
- Advancements in AI and generative AI are pushing the boundaries of what was once considered far-fetched in the computing sector.
- LLMs bridge the gap between human understanding and machine learning to offer a better content output.
- LLMs continue to improve their ability to provide logical and trustworthy responses across many complex knowledge sectors.
- Pre-trained language representation models (LRM) can be fine-tuned for specific tasks like text classification and language generation.
- LLMs are highly beneficial for problem-solving and helping businesses with communication-related tasks.
- LLMs can generate natural-sounding translations across multiple languages, enable code and text generation, and perform tasks with minimal training examples or without any training at all.
- Large language models have challenges and limitations that may affect their efficacy and real-world usefulness.
- Duke University's specialized course teaches students about developing, managing, and optimizing LLMs across multiple platforms, including Azure, AWS, and Databricks.
Read Full Article
12 Likes
Medium
1w
29
Image Credit: Medium
Achieving Reliable Model Performance with cross_val_score
- Machine learning models are not always stable and we have to evaluate the stability of the machine learning model.
- Cross-Validation is a technique used to assess how well our Machine learning models perform on unseen data.
- Cross-Validation is a resampling technique with the fundamental idea of splitting the dataset into 2 parts- training data and test data.
- One of the major advantages of the Hold Out method is that it is computationally inexpensive compared to other cross-validation techniques.
- The Leave One Out Cross-Validation method gives unbiased estimates but has an extremely high variance because the model is run 'n' times to test every observation.
- In K-Fold Cross-Validation, the mean of errors from all the iterations is calculated as the CV test error estimate.
- Stratified K-Fold CV will create K-Folds by preserving the percentage of sample for each class.
- Cross-validation is a powerful technique to ensure that your machine learning model generalizes well to unseen data.
- The sklearn.model_selection.cross_val_score function further simplifies this process, making it easier to implement cross-validation and obtain consistent evaluation metrics for your models.
- K-fold Cross Validation is typically performed using k=5 or k=10 as these values have been empirically shown to yield test error estimates that neither have high bias nor high variance.
Read Full Article
1 Like
Medium
1w
141
Image Credit: Medium
Apple's thinnest iPhone ever
- Apple is set to release its thinnest iPhone ever.
- The new iPhone will have a thickness of just 6 mm.
- Reports suggest that it will be called iPhone 17 Air or iPhone 17 Slim.
- The phone is expected to feature a 6.6-inch display and an Apple-made 5G modem.
Read Full Article
8 Likes
Analyticsindiamag
1w
397
Image Credit: Analyticsindiamag
When Chess Champion Gukesh Dommaraju Met Demis Hassabis
- Gukesh Dommaraju becomes the youngest World Chess Champion at 18.
- Demis Hassabis, CEO of DeepMind, praises Gukesh and the role of AI in chess.
- Google's sponsorship and AI contributions in chess highlight the integration of technology.
- AI cameras track players' focus and emotions in real time at the championship.
Read Full Article
23 Likes
Medium
1w
393
Image Credit: Medium
Agentic AI: Building Autonomous Systems from Scratch
- The rise of generative AI is transforming various industries, with a projected market value of over $65 billion in 2024.
- AI investment has surged, reaching $13.8 billion in 2024, as businesses incorporate AI into their strategies and systems.
- Large Language Models (LLMs) have paved the way for Agentic AI, a framework where autonomous agents collaborate to execute complex tasks.
- A demonstration showcases the development of a multi-agent system, integrating specialized agents for enhanced workflows.
Read Full Article
23 Likes
Analyticsindiamag
1w
350
Image Credit: Analyticsindiamag
The Future of Research is Agentic
- Google has launched a Deep Research tool that works on the Gemini bot and can search hundreds of websites within minutes.
- Gemini models are moving into agent-based areas that reason, plan, and act in the real world.
- The new Deep Research feature from Google is considered one of the most appropriate uses of AI and is impressive.
- Collaboration between AI labs and publishers will enable ethical access to paywalled content, supporting agentic research, knowledge sharing, and innovation.
Read Full Article
21 Likes
Medium
1w
247
Image Credit: Medium
Ertuğrul Ghazi, an historical figure of the 13th century, is also famous as the father of Osman I…
- Ertuğrul Ghazi is a historical figure known as the father of Osman I.
- His early and later life is still obscure, with limited information available.
- Ertuğrul played a role in the Seljuk Sultanate and contributed to their victory over the Byzantines.
- His legacy has been immortalized in the popular Turkish TV series 'Diriliş.'
Read Full Article
14 Likes
Dev
1w
29
Image Credit: Dev
Understanding Dijkstra's Algorithm: From Theory to Implementation
- Dijkstra's algorithm is a classic pathfinding algorithm used in graph theory to find the shortest path from a source node to all other nodes in a graph.
- It was proposed by Edsger W. Dijkstra in 1956 and remains one of the most widely used algorithms in computer science.
- Dijkstra's algorithm is a greedy algorithm designed to find the shortest paths from a single source node in a weighted graph with non-negative edge weights.
- The article provides core concepts of Relaxation, Priority Queue and Greedy Approach used in the Dijkstra algorithm.
- The relaxation ensures that the shortest distances are found by progressively updating the distance when a shorter path is found. The priority queue always dequeues the node with the smallest tentative distance, and the Greedy Approach processes nodes in non-decreasing order of their shortest distances.
- The correctness of Dijkstra’s algorithm has been proven through an inductive proof.
- The article also provides a JavaScript implementation of the Dijkstra algorithm.
- The different types of priority queues for Dijkstra's implementation, like Simple Array, Binary Heap, Binomial Heap, Fibonacci Heap, and Pairing Heap, have been discussed.
- The article concludes with the key points of Dijkstra’s algorithm and resources to explore the topic.
- Dijkstra’s algorithm helps in finding shortest paths in graphs, and it’s widely used in networking, routing, and other applications.
Read Full Article
1 Like
Analyticsindiamag
1w
81
Image Credit: Analyticsindiamag
Why Traditional SaaS is Under Threat
- Klarna, a fintech unicorn valued at $46bn, is ending its service provider relationships with Salesforce and Workday as part of a major internal overhaul driven by AI initiatives.
- A number of firms are opting to move away from its previous SaaS providers in favour of building in-house solutions.
- Many feel that with the advent of GenAI coding tools like GitHub Copilot and Anthropic’s Claude, one can expect software development to become cheaper and the job market for coders to evolve, creating a more accessible environment for talent, although at lower price points.
- Developing in-house AI involves higher initial costs due to infrastructure, talent acquisition, and maintenance expenses.
- However, in the long run, SaaS costs can add up as subscription fees scale with use.
- “The choice between in-house AI development and SaaS adoption depends heavily on an organisation’s specific situation, goals, and resources,” said Pradeep Sanyal, AI and data leader at a global tech consulting company.
- Notably, all SaaS companies, including Salesforce and Oracle, are stepping up efforts to integrate AI solutions into their offerings.
- On the other hand, NVIDIA chief Jensen Huang believes that SaaS is sitting on a goldmine.
- Hybrid approach is becoming popular as companies are using SaaS to initiate quickly and fill gaps while building in-house capabilities for their most critical, differentiating AI needs.
- B2B SaaS companies, which are taking a lot of data from businesses, will have a gloomy future.
Read Full Article
4 Likes
For uninterrupted reading, download the app