Latest news with #AWSBedrock
TECHx
28-04-2025
- Business
- TECHx
Dataiku Launches AI Agents to Boost Enterprise Innovation
Home » Emerging technologies » Artificial Intelligence » Dataiku Launches AI Agents to Boost Enterprise Innovation Dataiku, The Universal AI Platform™, has announced the launch of new AI agents. These capabilities are designed to help companies create, control, and scale AI agents for business applications. Over the past year, the adoption of GenAI and AI agents has accelerated. Today, more than 20% of Dataiku's customers use the platform to integrate GenAI into their workflows. Many customers have over 1,000 active use cases. However, rapid growth has created challenges. Many businesses deploy agents without IT control. This leads to inconsistent quality, security risks, and unmanaged sprawl. Dataiku addresses these issues by embedding agents inside a trusted, governed system. According to Florian Douetteau, co-founder and CEO of Dataiku, companies are now rethinking decades of enterprise applications. AI-native apps are becoming critical. These apps require a combination of analytics, predictive models, and agents. Only a platform like Dataiku can deliver all three together. As organizations move from pilot projects to full deployment, governance becomes essential. Without it, agent performance can decline, and technical debt can grow. Dataiku is building the infrastructure needed for centralized creation, performance monitoring, and orchestration. The platform supports both no-code and full-code agent development. Visual agent is ideal for business users. Code agent is built for developers. Both options come with governance features like managed tools, a GenAI registry, and risk validation workflows. Security is also a top priority. Dataiku's LLM Mesh architecture manages model access across vendors such as OpenAI, Anthropic, and Mistral. It also supports cloud services like Azure, AWS Bedrock, and Google Gemini, as well as self-hosted open-source models. Dataiku offers flexible guardrails with Safe Guard. It also provides Agent Connect to unify agent access across teams, allowing for easier orchestration. Continuous optimization is crucial. Agents can fail or drift over time. Dataiku's Trace Explorer, Quality Guard, and Cost Guard tools help monitor decisions, measure performance, and control costs. The platform connects AI agents to existing business workflows. It supports major cloud and data environments, including Snowflake, Databricks, Microsoft, AWS, and Google. This integration drives faster ROI while reducing complexity. AI agents with Dataiku are available today. Companies can now strengthen their AI strategies with greater control, security, and scalability.
Yahoo
25-04-2025
- Business
- Yahoo
Amazon Stock Is Down 28%. Should You Buy the Dip Before May 1?
On April 2, President Trump announced plans to enact various tariffs on imported goods from practically all of America's trading partners. Amazon (NASDAQ: AMZN) sources products from all over the world for its e-commerce platform, so it's facing the prospect of significantly higher costs for the products sold on its site. Whether it chooses to absorb those costs or pass them on to its customers, the tariffs have real potential to put a serious dent in its margins. Fortunately, Amazon is a diversified technology conglomerate. Not all its segments are directly affected by the trade tensions. Its Amazon Web Services (AWS) cloud platform primarily sells digital services, which aren't subject to traditional tariffs. That's great news because AWS accounts for most of the company's profits. Where to invest $1,000 right now? Our analyst team just revealed what they believe are the 10 best stocks to buy right now. Continue » Given the tariff turmoil, Amazon stock is trading down 28% from its recent all-time high. The company is scheduled to release its financial results for the first quarter of 2025 on May 1, which could be a positive catalyst. Should investors buy the dip ahead of the report? AWS offers hundreds of cloud solutions to help businesses transition into the digital age, whether they need basic data storage, video streaming capabilities, or complex software development tools. It's the largest platform of its kind in the world, and it's using its immense scale to move into artificial intelligence (AI), which Amazon CEO Andy Jassy calls a once-in-a-lifetime business opportunity. AWS is tackling the three core layers of AI -- data center infrastructure, large language models (LLMs), and software: AWS has designed its own data center chips, called Trainium, to help reduce its customers' AI training costs by up to 40% from those charged by third-party suppliers (like Nvidia). It has also developed its own family of LLMs called Nova, which can offer cost savings of 75% compared to models from other vendors on the AWS Bedrock platform. To cover the third layer, AWS has built a powerful virtual assistant called "Q." It can write programming code to accelerate software projects, and it can help businesses analyze their internal data to uncover new opportunities to generate revenue. AWS generated a record $107.5 billion in total revenue during 2024, which represented just 16.8% of Amazon's total revenue of $637.9 billion. However, AWS is highly profitable, so it accounted for more than half of the organization's entire operating income of $68.6 billion. The platform's quarterly revenue growth accelerated to 19% in the early stages of last year and stayed there. On May 1, Wall Street will want to see whether AWS built on that momentum in the first quarter of 2025 and whether management says its growing portfolio of AI services was a major contributor. E-commerce remains Amazon's single largest source of revenue, but it operates on razor-thin profit margins, so the company has focused on improving its efficiency over the last couple of years. It divided its U.S. logistics network into eight distinct regions in 2023, which allows the company to stock different products in different fulfillment centers depending on their popularity in specific geographic locations. This means orders travel shorter distances to reach customers, which lowers costs and improves delivery times. Those adjustments contributed to significant earnings growth for Amazon last year. Unfortunately, tariffs threaten to undo its progress. As things stand today, tariffs will increase the cost of every product Amazon imports into the U.S. by at least 10%, and some products coming from China, specifically, are set to become a staggering 245% more expensive. Wall Street will be eager to hear how Andy Jassy plans to navigate this issue, and the performance of Amazon stock in the short term could hinge on whether analysts are satisfied with his strategy. Nevertheless, Wall Street's consensus estimate (as provided by Yahoo! Finance) suggests Amazon could deliver $1.36 in earnings per share (EPS) during Q1, which would be a solid 38.7% increase from the year-ago period. Simply put, AWS is likely to support the company's earnings amid the global trade tensions, as are segments like digital advertising and video streaming, which aren't directly subject to tariffs. The 28% dip in Amazon stock from its record high has created an opportunity for investors to buy it at a very attractive valuation. It now trades at a price-to-earnings (P/E) ratio of just 31.1, which is a steep discount to its five-year average of 83. I'm not suggesting it will get back there because 83 is very high, but Amazon has always traded at a big premium to the Nasdaq-100 index, which currently sits at a P/E ratio of 27.1. If we look ahead to 2026, Wall Street expects Amazon to deliver $7.52 in EPS, which places the stock at a forward P/E ratio of just 22.9. Therefore, the stock would have to climb by 35.8% by the end of next year just to maintain its current P/E ratio of 31.1: Simply put, the biggest reason to take a long-term position in Amazon stock today is its current valuation, rather than what might come out of the company's Q1 report on May 1. After all, Amazon has an incredible track record of success, which is why its stock has soared by a staggering 191,000% since it went public in 1997. One single quarter is unlikely to change its trajectory, so investors could earn a positive return over the long run whether they buy it ahead of next Thursday or wait until after the Q1 results are released. Before you buy stock in Amazon, consider this: The Motley Fool Stock Advisor analyst team just identified what they believe are the for investors to buy now… and Amazon wasn't one of them. The 10 stocks that made the cut could produce monster returns in the coming years. Consider when Netflix made this list on December 17, 2004... if you invested $1,000 at the time of our recommendation, you'd have $561,046!* Or when Nvidia made this list on April 15, 2005... if you invested $1,000 at the time of our recommendation, you'd have $606,106!* Now, it's worth noting Stock Advisor's total average return is 811% — a market-crushing outperformance compared to 153% for the S&P 500. Don't miss out on the latest top 10 list, available when you join . See the 10 stocks » *Stock Advisor returns as of April 21, 2025 John Mackey, former CEO of Whole Foods Market, an Amazon subsidiary, is a member of The Motley Fool's board of directors. Anthony Di Pizio has no position in any of the stocks mentioned. The Motley Fool has positions in and recommends Amazon and Nvidia. The Motley Fool has a disclosure policy. Amazon Stock Is Down 28%. Should You Buy the Dip Before May 1? was originally published by The Motley Fool Sign in to access your portfolio
Yahoo
20-04-2025
- Business
- Yahoo
Warren Buffett's Berkshire Hathaway Invests In These 3 AI Stocks — Why You Should Too
While Warren Buffett's Berkshire Hathaway hasn't been able to evade the market misfortunes stymieing stock valuations over the course of recent days (Class A shares shed 2.25% of value as of 1 p.m. ET on April 10, while Class B shares lost 2.8%), the Oracle of Omaha's guidance remains invaluable for long-term investors. Berkshire Hathaway holds long positions on these three prominent AI-adjacent or AI-related stocks, and they may also be top picks for those considering an addition to their portfolios. Check Out: Read Next: According to the Financial Post,1 Berkshire Hathaway picked up 1.3 million shares in Domino's during Q3 2024, worth about $550 million. Trading at about $428 per share in mid-October, and $446 as of April 10 — despite market turmoil — it's likely that Buffett's call was the correct one. Domino's utilizes AI technology in a variety of ways, including implementing Microsoft's Azure platform for its agentic AI assistants and predictive ordering models to get pizzas out the door more efficiently. Learn More: While Berkshire Hathaway has been shedding Amazon shares at a profit over the course of the past few years, it still maintains a sizable position in the retail giant, per Stockcircle.2 Buffett's multinational conglomerate holds about 10 million shares as of this writing, worth approximately $1.91 billion. Amazon Web Services is one of the leading names in the cloud business, responsible for building massive data centers leveraging Nvidia hardware to make leaps in the AI space. AI tools such as Claude and Llama also have a home within the AWS Bedrock platform, making Amazon a key player in the nascent artificial intelligence industry. Coca-Cola has long been a favorite of Buffett's, with the Oracle of Omaha first buying in to the beverage business leader way back in 1988. Since then, it has been a textbook example of Buffett buying — and holding — solid long-term investments with growth potential. But despite Coca-Cola striking a deal to implement the Microsoft Azure AI platform to streamline operations, this buy may be more of an anchor or hedge in one's portfolio rather than a significant short-term profit opportunity. Nonetheless, having a guaranteed performer in one's holdings can reduce volatility. More From GOBankingRates 6 Used Luxury SUVs That Are a Good Investment for Retirees4 Affordable Car Brands You Won't Regret Buying in 20257 Overpriced Grocery Items Frugal People Should Quit Buying in 20255 Types of Vehicles Retirees Should Stay Away From Buying Sources Financial Post, 'Warren Buffett's Berkshire buys stakes in Domino's Pizza and Pool Corp.' (Nov. 15, 2024) ︎ Stockcircle, 'Berkshire Hathaway's Stake.' ︎ This article originally appeared on Warren Buffett's Berkshire Hathaway Invests In These 3 AI Stocks — Why You Should Too Sign in to access your portfolio
Yahoo
19-04-2025
- Business
- Yahoo
Warren Buffett's Berkshire Hathaway Invests In These 3 AI Stocks — Why You Should Too
While Warren Buffett's Berkshire Hathaway hasn't been able to evade the market misfortunes stymieing stock valuations over the course of recent days (Class A shares shed 2.25% of value as of 1 p.m. ET on April 10, while Class B shares lost 2.8%), the Oracle of Omaha's guidance remains invaluable for long-term investors. Berkshire Hathaway holds long positions on these three prominent AI-adjacent or AI-related stocks, and they may also be top picks for those considering an addition to their portfolios. Check Out: Read Next: According to the Financial Post,1 Berkshire Hathaway picked up 1.3 million shares in Domino's during Q3 2024, worth about $550 million. Trading at about $428 per share in mid-October, and $446 as of April 10 — despite market turmoil — it's likely that Buffett's call was the correct one. Domino's utilizes AI technology in a variety of ways, including implementing Microsoft's Azure platform for its agentic AI assistants and predictive ordering models to get pizzas out the door more efficiently. Learn More: While Berkshire Hathaway has been shedding Amazon shares at a profit over the course of the past few years, it still maintains a sizable position in the retail giant, per Stockcircle.2 Buffett's multinational conglomerate holds about 10 million shares as of this writing, worth approximately $1.91 billion. Amazon Web Services is one of the leading names in the cloud business, responsible for building massive data centers leveraging Nvidia hardware to make leaps in the AI space. AI tools such as Claude and Llama also have a home within the AWS Bedrock platform, making Amazon a key player in the nascent artificial intelligence industry. Coca-Cola has long been a favorite of Buffett's, with the Oracle of Omaha first buying in to the beverage business leader way back in 1988. Since then, it has been a textbook example of Buffett buying — and holding — solid long-term investments with growth potential. But despite Coca-Cola striking a deal to implement the Microsoft Azure AI platform to streamline operations, this buy may be more of an anchor or hedge in one's portfolio rather than a significant short-term profit opportunity. Nonetheless, having a guaranteed performer in one's holdings can reduce volatility. More From GOBankingRates 6 Used Luxury SUVs That Are a Good Investment for Retirees4 Affordable Car Brands You Won't Regret Buying in 20257 Overpriced Grocery Items Frugal People Should Quit Buying in 20255 Types of Vehicles Retirees Should Stay Away From Buying Sources Financial Post, 'Warren Buffett's Berkshire buys stakes in Domino's Pizza and Pool Corp.' (Nov. 15, 2024) ︎ Stockcircle, 'Berkshire Hathaway's Stake.' ︎ This article originally appeared on Warren Buffett's Berkshire Hathaway Invests In These 3 AI Stocks — Why You Should Too Sign in to access your portfolio
Forbes
27-03-2025
- Business
- Forbes
Your Essential Primer On Large Language Model Agent Tools
Mohammad Adnan is Principal Engineer & AI trailblazer at Intuit, driving next-gen automation for small business; ex‑AWS leader. getty Having spent years building and scaling artificial intelligence and machine language (AI/ML) solutions at AWS Bedrock and now at Intuit, I've witnessed firsthand the incredible advancements in large language models (LLMs). Although initial excitement often revolves around single-turn interactions, the real power unlocks when we orchestrate these models to tackle complex tasks through intelligent, multistep processes. This is where AI agents come into play. For example, if you wanted to plan a multi-city trip with specific budget and activity constraints, an AI agent powered by these frameworks could automate the entire process—from researching flights to managing your budget—something a simple prompt can't achieve. In this article, I'll share my experience navigating the landscape of various agent frameworks through a practical comparison of several popular LLM agent tools. We'll explore their unique strengths and weaknesses and how you can leverage them within your own use cases. LangChain is your go-to if you need a highly flexible and extensively integrated framework. Its massive, active community provides a wealth of templates, plugins and prompt-chaining strategies. The sheer number of available integrations means you can connect your AI agent to virtually any API or data source. Plus, its robust memory management allows you to tailor how your agent retains information across multiple steps. Use LangChain when: You have complex tasks requiring integration with diverse tools and data sources, need fine-grained control over memory management and want to leverage a large and supportive community. However, the sheer breadth of LangChain can be overwhelming for beginners, leading to a steeper learning curve. Debugging intricate prompt chains can also be challenging. And although cost-effective, scaling can require significant engineering effort. Consider another option if: You're looking for a simpler, more visually oriented approach or are just starting your journey with AI agents. Typical Use Case: LangChain excels at building sophisticated product support chatbots that can consult internal documentation, summarize it and engage in multiturn conversations to refine answers based on user queries. If clarity and simplicity are your priorities, LangGraph is an excellent choice. Its node-based design provides a visual representation of your agent's workflow, making it easy to understand and manage. Defining discrete "nodes" for each step offers a more intuitive approach compared to code-heavy chaining. Use LangGraph when: You prefer a visual, easy-to-understand way to build AI agent workflows, are working on smaller applications or value a clear pipeline view. It's a great starting point for teams new to agent frameworks. Although its simplicity is a strength, LangGraph might lack some of the more advanced features and extensive integrations found in LangChain. For very complex scenarios with numerous conditional branches or specialized tools, it might require more custom development. Consider other options if: You anticipate needing a vast array of pre-built integrations or highly intricate workflow logic right out of the box. Typical Use Case: LangGraph is ideal for building small to medium-scale question-answering applications where you need a clear, step-by-step flow that's easy for developers to trace and understand. CrewAI (Python) For enterprise-level applications requiring collaboration among multiple specialized AI agents, CrewAI is the framework to consider. Its focus on multi-agent orchestration, complete with role-based access control, logging, and monitoring, makes it suitable for complex organizational needs. The ability for agents to share results and escalate tasks enables sophisticated problem-solving. Use CrewAI when: You need to build applications with multiple interacting agents, have enterprise-level security and compliance requirements and need to manage complex, multistep workflows involving different specialized AI roles. However, setting up and managing the interactions between multiple agents in CrewAI can introduce complexity. Debugging issues across a team of agents might also be more involved. Consider other options if: You're building single-agent applications or have simpler collaboration needs. Typical Use Case: CrewAI is well-suited for regulated industries like finance, where you might need multiple agents to parse legal documents, check for policy risks and compile final summaries while maintaining detailed access logs and ensuring compliance. SpringAI (Java) If your organization is heavily invested in the Java ecosystem, Spring AI offers a seamless way to integrate LLM capabilities into your existing applications. Its tight integration with Spring Boot and familiar Spring patterns makes it a natural choice for Java developers. Use Spring AI when: Your primary development language is Java and you want to easily embed LLM functionalities into your Spring Boot applications without switching languages. Spring AI's primary limitation is its focus on Java. It's not the right choice for teams using other languages. Additionally, its built-in agent orchestration capabilities are currently less advanced compared to Python-based frameworks. Consider other options if: Your team primarily works with Python or you require more sophisticated, out-of-the-box agent orchestration features. Typical Use Case: A healthcare firm maintaining Java microservices for patient data can use Spring AI to quickly add LLM-driven summarization or question-answer features to their existing endpoints. AutoGen (Python) AutoGen is your go-to framework when the primary goal is to generate and refine high-quality code. Its unique coder-reviewer agent workflow leads to iterative improvements, reducing debugging time. Use AutoGen when: Your main application involves generating code and you value the automated review and refinement process to improve code quality and reduce errors. The iterative code generation process can sometimes be slower than a single-pass approach. It also requires careful configuration to ensure the coder and reviewer agents work effectively together. Consider other options if: Your application doesn't primarily involve code generation or you need rapid, single-step outputs. Typical Use Case: A development team building Python scripts for data cleaning can use AutoGen to have a "coder" agent propose an initial solution, followed by a "reviewer" agent that identifies potential issues and prompts revisions. Bedrock Agent (AWS) If you're deeply embedded in the AWS ecosystem and want a hassle-free way to build intelligent applications, Bedrock Agents offers a fully managed experience. It gives you easy access to a variety of powerful language models and takes care of the underlying infrastructure, letting you focus on building your AI-powered solutions. Use Bedrock Agents when: You're all-in on AWS and prioritize a managed service for building AI agents with diverse foundation models and integration with other AWS tools. However, remember that being tightly integrated with AWS means you're also tied to their platform. You'll have less direct control over the nuts and bolts, and costs can add up if your usage is high. Consider other options if: You prefer the flexibility and control of open-source frameworks or are looking for cost-optimization strategies that might be available outside of a fully managed environment. Typical Use Case: A large e-commerce company already running entirely on AWS could use Bedrock Agents to create a sophisticated product recommendation system that not only understands what customers are asking but also dives into product details, compares options and offers personalized suggestions. Conclusion Choosing the right AI agent framework is a crucial step in building intelligent and efficient applications. Although each framework offers unique advantages, remember that experimentation and hands-on experience are key to unlocking the full potential of AI agents. Which framework are you most excited to explore, and what innovative applications do you envision building? Forbes Technology Council is an invitation-only community for world-class CIOs, CTOs and technology executives. Do I qualify?
