Trump calls on Fed Governor Cook to resign
Representatives for Cook could not be immediately reached for comment on the allegations posted by FHFA Director Bill Pulte on X earlier on Wednesday.
Hashtags
Try Our AI Features
Explore what Daily8 AI can do for you:
Comments
No comments yet...
Related Articles
Geeky Gadgets
27 minutes ago
- Geeky Gadgets
Context Engineering vs Prompt Engineering : The Secret Sauce to Smarter AI
What if the key to unlocking truly intelligent AI isn't just about asking the right questions, but about building the perfect environment for those questions to thrive? While much of the conversation around AI optimization has focused on prompt engineering—the art of crafting precise instructions for large language models (LLMs)—a quieter revolution is reshaping the field. Enter context engineering, a broader, system-level approach that equips AI with the tools, memory, and data it needs to perform at its best. Together, these methodologies are redefining what's possible, allowing AI systems to tackle complex, real-world challenges with unprecedented precision and adaptability. In this exploration, the IBM Technology team unravel the nuanced interplay between prompt engineering and context engineering, revealing how their combined power is shaping the future of AI. You'll discover how retrieval-augmented generation (RAG)* and state management are transforming AI into smarter, more dynamic systems. Along the way, we'll examine why context engineering is emerging as the unsung hero of AI optimization, offering solutions to challenges that prompt engineering alone cannot solve. Whether you're building AI for customer support, financial analysis, or personalized recommendations, understanding this synergy could be the key to unlocking its full potential. After all, the smartest AI isn't just well-instructed, it's well-equipped. Prompt vs. Context Engineering What is Prompt Engineering? Prompt engineering is the art and science of designing input instructions that guide how LLMs interpret tasks and generate responses. By carefully structuring prompts, you can influence the AI's behavior, improve the relevance of its outputs, and tailor its responses to specific needs. This methodology is particularly valuable for making sure that AI systems deliver accurate and contextually appropriate results. Key techniques in prompt engineering include: Role Assignment: Defining the AI's role, such as a teacher, assistant, or analyst, to shape its tone, style, and approach to the task. Defining the AI's role, such as a teacher, assistant, or analyst, to shape its tone, style, and approach to the task. Few-Shot Learning: Providing examples within the prompt to help the AI understand the desired output format and context. Providing examples within the prompt to help the AI understand the desired output format and context. Chain-of-Thought Prompting: Encouraging the AI to use step-by-step reasoning, which enhances its ability to solve complex problems. Encouraging the AI to use step-by-step reasoning, which enhances its ability to solve complex problems. Constraint Setting: Specifying rules or limitations to ensure the AI's outputs meet specific criteria, such as tone, length, or format. For example, if you ask an AI to summarize a lengthy report, a well-crafted prompt might specify the desired length, tone, and key points to include. This level of precision directly impacts the quality and usefulness of the AI's response, making prompt engineering a critical skill for optimizing AI performance. What is Context Engineering? Context engineering takes a broader approach, focusing on creating an environment where the AI has access to the tools, memory, and data it needs to make informed decisions. Unlike prompt engineering, which deals with task-specific instructions, context engineering ensures that the AI system is equipped to handle multi-step, complex tasks by integrating external resources and managing internal states. Key components of context engineering include: Memory Management: Organizing short-term and long-term memory to improve continuity and personalization. Short-term memory might summarize ongoing conversations, while long-term memory could store user preferences in a vector database for future interactions. Organizing short-term and long-term memory to improve continuity and personalization. Short-term memory might summarize ongoing conversations, while long-term memory could store user preferences in a vector database for future interactions. State Management: Tracking progress in multi-step tasks to maintain continuity and avoid redundant or conflicting actions. Tracking progress in multi-step tasks to maintain continuity and avoid redundant or conflicting actions. Retrieval-Augmented Generation (RAG): Dynamically extracting relevant information from external knowledge sources using hybrid search techniques, such as combining keyword search with vector-based similarity. Dynamically extracting relevant information from external knowledge sources using hybrid search techniques, such as combining keyword search with vector-based similarity. Tool Integration: Allowing the AI to interact with external systems, such as APIs, databases, or live data feeds, to expand its capabilities and deliver more comprehensive results. For instance, an AI tasked with planning a vacation might use context engineering to retrieve real-time flight data, access user preferences stored in memory, and interact with booking APIs. This integration allows the AI to deliver seamless, personalized recommendations that align with the user's needs and preferences. Context Engineering vs. Prompt Engineering: Smarter AI with RAG & Agents Watch this video on YouTube. Here are more guides from our previous articles and guides related to Large Language Models (LLMs) that you may find helpful. How Prompt and Context Engineering Work Together Prompt engineering and context engineering are not isolated methodologies; they are complementary and interdependent. While prompt engineering refines the instructions given to the AI, context engineering ensures the system has the necessary resources to execute those instructions effectively. Together, they create a feedback loop that enhances the AI's overall performance. Consider an AI agent managing a travel booking. A well-designed prompt might outline the user's requirements, such as destination, budget, and travel dates. Context engineering, on the other hand, ensures the AI can retrieve relevant flight options, check hotel availability, and remember the user's preferences from previous interactions. By combining these strategies, the AI can deliver accurate, tailored results that meet the user's expectations. This synergy is particularly valuable in applications requiring both precision and adaptability. Whether it's assisting with customer support, conducting financial analysis, or managing supply chains, the interplay between prompt and context engineering enables AI systems to handle real-world challenges with greater efficiency and accuracy. Challenges and Opportunities Both prompt engineering and context engineering present unique challenges. Prompt engineering requires a deep understanding of LLM behavior to craft effective instructions that yield the desired outcomes. Context engineering, on the other hand, demands robust infrastructure to manage memory, track states, and retrieve data efficiently. These challenges can be resource-intensive, particularly for organizations without advanced technical expertise. However, advancements in supporting technologies are making these methodologies more accessible. Tools like vector databases, hybrid search techniques, and APIs are simplifying the implementation of context engineering, while improved LLMs are enhancing the effectiveness of prompt engineering. The integration of retrieval-augmented generation (RAG) further expands the capabilities of AI systems by allowing them to access dynamic, up-to-date knowledge sources. This is especially valuable in fields like healthcare, finance, and customer service, where real-time information is critical. As these methodologies continue to evolve, they open up new opportunities for innovation. By combining the precision of prompt engineering with the adaptability of context engineering, organizations can develop AI systems capable of tackling increasingly complex tasks, from automating workflows to delivering personalized user experiences. The Future of AI Optimization Prompt engineering and context engineering represent two essential pillars of AI optimization. While prompt engineering focuses on how you communicate with LLMs, context engineering builds the ecosystem that supports intelligent decision-making. Together, these methodologies enable the creation of AI agents that are not only smarter but also more adaptable and capable of handling real-world challenges. As AI technology advances, the interplay between these approaches will continue to drive innovation. By using the strengths of both prompt and context engineering, you can unlock the full potential of AI, creating systems that deliver accurate, efficient, and personalized results across a wide range of applications. The future of AI lies in this synergy, where precision meets adaptability to redefine what intelligent systems can achieve. Media Credit: IBM Technology Filed Under: AI, Top News Latest Geeky Gadgets Deals Disclosure: Some of our articles include affiliate links. If you buy something through one of these links, Geeky Gadgets may earn an affiliate commission. Learn about our Disclosure Policy.
The Independent
27 minutes ago
- The Independent
Reporter says former aide to Eric Adams gave her a potato chip bag filled with cash
A longtime adviser to New York City Mayor Eric Adams, who resigned from his administration while under FBI scrutiny, reportedly gave a journalist a potato chip bag filled with cash on Wednesday following a campaign event. Winnie Greco, a former aide, presented the unusual gift to Katie Honan, a reporter for the local news site The City. The bag reportedly contained a red envelope with a £100 note and several £20 notes. Ms Greco's lawyer later insisted the offering was not an attempted bribe. Ms Honan had previously scrutinised Ms Greco's conduct as a major fundraiser for Mayor Adams within the Chinese American community. Greco's attorney, Steven Brill, told The Associated Press that the situation was being 'blown out of proportion." 'This was not a bag of cash,' Brill wrote in an email. 'In the Chinese culture, money is often given to others in a gesture of friendship and gratitude. And that's all that was done here. Winnie's intention was born purely out of kindness.' Asked why Greco wanted to make such a gesture to Honan, Brill said, 'She knows the reporter and is fond of her.' The City said it interviewed Greco later Wednesday and she apologized, saying she made 'a mistake.' 'I'm so sorry. It's a culture thing. I don't know. I don't understand. I'm so sorry. I feel so bad right now,' Greco said, according to The City. In response to the report of the bag filled with cash, Adams' reelection campaign said it had suspended Greco from further work as an unpaid volunteer and that Adams had no prior knowledge of Greco's actions. The City reported Greco had texted Honan to meet her inside a Whole Foods store after they both attended the opening of Adams' campaign headquarters in Harlem. When given the chip bag, Honan at first thought Greco was just giving her a snack and said she could not accept it but Greco insisted, according to the report. Honan left and later discovered the money, then called Greco and told her she could not accept it and asked to give it back. Greco said they could meet later but then stopped responding, the report said. Greco later called The City back and asked them not to do a story, saying 'I try to be a good person," the news outlet reported. A City Hall spokesperson declined to comment Wednesday night. An Adams campaign aide, Todd Shapiro, said Greco holds no position in the campaign. 'We are shocked by these reports,' Shapiro said. 'Mayor Adams had no prior knowledge of this matter. He has always demanded the highest ethical and legal standards, and his sole focus remains on serving the people of New York City with integrity.' A text message sent to a phone number listed in public records for Greco was not immediately returned Wednesday night. Since she resigned as Adams' director of Asian affairs last fall, Greco has occasionally been seen at Adams campaign events. Before her resignation, Greco had served as Adams' longtime liaison with the city's Chinese American community. She was also a prolific fundraiser for Adams' campaigns. In February of 2024, federal agents searched two properties belonging to Greco. Authorities didn't explain what the investigation was about, and Greco has not been charged with committing a crime, but she was a number of close aides to Adams who resigned or were fired amid the federal scrutiny. The City has reported extensively on the investigation and Greco's conduct, including a campaign volunteer's allegations that Greco had promised to get him a city job if he helped renovate her home. A separate federal investigation into Adams led to a 2024 indictment accusing the mayor of accepting illegal campaign contributions and travel discounts from a Turkish official and others — and returning the favors by, among other things, helping Turkey open a diplomatic building without passing fire inspections. A federal judge dismissed the case in April after the Justice Department ordered prosecutors to drop the charges, arguing that the case was interfering with the mayor's ability to aid President Donald Trump 's crackdown on illegal immigration.
The Independent
27 minutes ago
- The Independent
WH Smith shares tumble after lowering profit outlook on US accounting error
WH Smith has warned its yearly profits will be lower than previously expected due to an accounting error in the US, sending its share price tumbling. Shares in the London-listed travel retailer were down by about a third on Thursday morning. WH Smith said it discovered its trading profit in North America had been overstated by about £30 million, when reviewing its finances. This was because of an issue in how it calculated the amount of supplier income it received – leading it to be recognised too early. It means the group is now expecting a trading profit for the US of about £25 million for the year to August – a cut from the previous £55 million forecast. As a result, the company lowered its outlook for annual pre-tax profits to around £110 million. The London-listed business incorporates its travel locations, such as shops in airports, train stations and hospitals, which total about 1,300 around the world. Whereas the high street chain of about 480 shops sold to Hobbycraft owner Modella Capital in June. As part of the deal, the WH Smith name will disappear from British high streets and be replaced by brand TGJones. The travel locations were not included in the sale and will not be changing. WH Smith shares were down by about 35% in early trading on Thursday.
