Latest news with #AgentMode
Business Wire
2 days ago
- Business
- Business Wire
Postman Unveils Agent Mode: AI-Native Development Revolutionizes API Lifecycle
LOS ANGELES--(BUSINESS WIRE)-- POST/CON — Postman, the world's leading API collaboration platform, today announced Agent Mode, a breakthrough AI-native assistant that delivers real productivity gains across the entire API lifecycle. Available directly within the Postman platform, Agent Mode understands developer intent and executes real tasks – designing, testing, documenting, and monitoring APIs – based on simple natural language input. It compresses weeks of manual effort into hours, enabling teams to ship faster with fewer handoffs and less friction. Organizations are turning to AI for software development at an increasingly rapid pace. According to IDC, this is driven by AI's ability to decrease the need for manual testing, resulting in improvements in test coverage, software usability, and code quality. Further, a McKinsey study showed that software developers can complete coding tasks up to twice as fast with generative AI, and these gains in developer productivity can lead to significant improvements in business outcomes, including faster time-to-market, higher profit margins, and increased customer satisfaction. As AI continues to reshape the software development landscape, Agent Mode positions Postman at the forefront of this industry evolution: from code-first tools to intent-first platforms. Where most tools offer copilots that assist with code suggestions, Postman offers a fully capable execution agent—connected to real APIs, aware of state, and embedded in a platform used by over 40 million developers worldwide. It actively manages the heavy lifting of API development, transforming the fragmented, manual nature of modern software work into fast, AI-powered flows. Agent Mode is an AI-powered assistant that turns requests into action—executing complex flows across Postman's collections, tests, environments, and documentation. Looking ahead, developers will also be able to build and reuse their own multi-step agents to automate common API tasks. These reusable agents will live in teams' workspace and scale productivity across functions and processes—from daily engineering tasks to broader business operations. 'Agent Mode signifies a major shift in how developers interact with tools, APIs, and the systems they build,' said Abhijit Kane, Co-Founder of Postman. 'With Agent Mode, Postman becomes the first platform to combine enterprise-grade API tooling with an intelligent agent capable of understanding business intent and delivering production-grade workflows on demand. It effectively removes the need for time-consuming prototyping and empowers teams to move from idea to working product in days - not quarters.' Agent Mode acts as a natural-language gateway to Postman's core capabilities: Collections and Environments: The agent creates, organizes, and updates collections with proper variable handling and reusable configurations. Automated Testing: Users can ask the agent to validate response schemas, create test cases, or simulate edge cases. Documentation: Once a flow is working, the agent can generate detailed API documentation that's immediately shareable across teams. Reusable Agents: Developers can build multi-step agents that automate repeatable API tasks—from engineering flows to business processes—all reusable and shareable within your workspace. Monitoring and Observability: Developers can set up monitors or define health checks, all through conversation. Team Collaboration: Everything created by the agent lives in Postman's collaborative workspace, instantly usable by all teammates, including product managers, quality assurance, support, and even non-technical members to reduce reliance on developers and expand the pool of 'builders' inside a company. To join the waitlist to learn more about Agent Mode, visit About Postman Postman is the world's leading API collaboration platform, used by more than 40 million developers and 500,000 organizations, including 98% of the Fortune 500. In a world that is increasingly embracing an API-first mindset, Postman is uniquely positioned to streamline API and AI application development and drive innovation through teamwork by transcending traditional testing and fostering team collaboration across the entire API development lifecycle — from creation to deployment. The company is headquartered in San Francisco with offices in Bangalore and Tokyo. Postman is privately held, with funding from Battery Ventures, BOND, Coatue, CRV, Insight Partners, and Nexus Venture Partners. Learn more at or connect with Postman on X via @getpostman.
Hans India
27-05-2025
- Business
- Hans India
Gemini I/O 2025: Ushering in the Era of World Model AI and Agentic Intelligence
With the unveiling of Gemini I/O 2025, Google signals a seismic shift in the AI landscape—ushering in a new age of world model AI and agentic intelligence. The implications go far beyond productivity: we're looking at AI systems that collaborate, learn, and adapt with human-like context. At the Gemini I/O 2025 event, Google introduced a vision of artificial intelligence that's not only more powerful, but fundamentally more human-aware. With the launch of the Gemini 2.5 Pro model and the debut of Agent Mode, the spotlight is now firmly on 'world models' and agentic AI—systems that don't just respond, but reason, plan, and evolve. 'Gemini I/O 2025 marks a decisive inflection point in the evolution of artificial intelligence—an era where we move from narrow, task-specific models to expansive 'World Models' that understand, reason, and act with context across environments,' said Jaspreet Bindra, Co-founder, AI&Beyond. 'This isn't just about better chatbots or smarter automation; it's the foundation for truly agentic intelligence—AI systems that can autonomously perceive goals, plan actions, and adapt to complex real-world dynamics." Bindra underscored that what sets world models apart is their multimodal capability fused with memory and reasoning—hallmarks of how humans interact with their environment. 'With Gemini's architecture, we are seeing the rise of AI that can collaborate, not just compute; that can anticipate, not just react,' he added. 'As we build toward a future where digital agents become trusted co-pilots in decision-making—from scientific discovery to enterprise productivity—we must also embed safety, alignment, and transparency at the core of these systems.' Echoing the sentiment, Mayank Maggon, Founder and CEO of TechChefz Digital, said, 'Gemini I/O 2025 marks a significant step forward in AI evolution—bridging the gap between intelligence and true autonomy.' He pointed out that Gemini 2.5 Pro is not merely a performance upgrade—it's a foundational leap. 'With the ability to process text, images, audio, and video simultaneously and handle up to 1 million tokens (soon expanding to 2 million), this opens up enterprise-grade use cases,' he said. Among these use cases: instant auditing of large codebases and compliance documents, deriving actionable insights from hours of meeting transcripts, and cross-referencing legal, financial, and product datasets in real time. The standout innovation, however, is the new Agent Mode—a framework for AI systems that not only execute tasks but learn from user behaviour over time. 'Imagine delegating your calendar management, project planning, or travel logistics to an AI that not only executes but learns your preferences. Or training the AI on specific workflows—like updating CRM entries or responding to RFPs—and letting it handle them independently,' said Maggon. This evolution transforms AI into a proactive, personalised assistant—capable of summarising customer feedback, drafting reports, or even generating creative content like design briefs, all based on internal communications and documents. The seamless integration of Gemini across Google's ecosystem—from Chrome and Gmail to Android Auto and smart home devices—ensures that this AI is always present, ambient, and contextually aware. 'Whether you're driving, in a meeting, or at home, AI support is now ambient and proactive,' Maggon added. 'At TechChefz Digital, we're actively exploring how next-gen AI like Gemini can augment internal workflows, enhance customer experience platforms, and power intelligent enterprise solutions.' As Jaspreet Bindra aptly concluded, 'Gemini I/O is more than a technical upgrade—it's a philosophical leap. The age of agentic, world-aware AI is no longer speculative—it's here, and it will transform every interface, workflow, and expectation we have from machines.'
Hans India
22-05-2025
- Business
- Hans India
Google's Vision: A Future Where AI Does the Googling for You
At this year's Google I/O, artificial intelligence wasn'tjust part of the story—it was the story. The tech giant unveiled a vision forthe future in which Google's AI doesn't just help you search, it does thesearching for you. The centre of this evolution is AI Mode in Google Search, nowrolling out across the U.S. This new mode transforms the traditional search barinto an intelligent, chatbot-like interface capable of understanding complexqueries and pulling together comprehensive, curated responses. It's a step awayfrom the familiar list of blue links and a leap toward an AI that actively doesyour online legwork. During the keynote, Google demonstrated how AI Mode couldplan a weekend getaway to Nashville for friends interested in food, music, andunique experiences. Instead of offering a basic search result, AI Mode createddynamic, themed suggestions such as 'restaurants good for foodies,' 'chill baratmosphere with live music,' and 'off-the-beaten-path attractions,' completewith a customised map and links to relevant sites. Behind this capability is what Google calls its 'queryfanout technique,' powered by a custom version of its Gemini model. Liz Reid,head of Google Search, explained: 'Now, under the hood, Search recognizes when a questionneeds advanced reasoning. It calls on our custom version of Gemini to break thequestion into different subtopics, and it issues a multitude of queriessimultaneously on your behalf... Search pulls together a response and checksits work to make sure it meets our high bar for information quality.' This means that what used to be several separate searchesare now bundled into one—executed and analyzed by Google's AI, which thendelivers a comprehensive response. The AI Mode interface even shows users howmany searches it's performing in the background. Later this summer, a new 'Deep Search' feature is slated toarrive within AI Mode. It builds on the same query fanout approach but scalesit significantly. According to Reid, Deep Search can 'issue dozens or evenhundreds of searches on your behalf' to produce even more in-depth answers. Another key player in Google's AI ecosystem is ProjectMariner, a behind-the-scenes tool that allows AI to perform complex web can juggle up to 10 operations simultaneously and features a 'Teach and Repeat'capability, allowing users to train the system to perform recurring tasks. Google is extending these functionalities to the Geminiapp's new Agent Mode, which also taps into Project Mariner. CEO Sundar Pichaigave a live example of how Agent Mode could help find an apartment in Austin byautomatically scanning listings on Zillow and surfacing the best matches. Additionally, Project Mariner will soon integrate into AIMode itself. Rajan Patel, VP of Engineering for Search, demonstrated how thetool could locate baseball tickets and provide a direct purchase button—allfrom within the search interface. What Google is proposing is a major transformation in how weinteract with the web. The company sees AI not just as an assistant but as areplacement for many of the routine search tasks users handle today. As Reidsummed it up: 'Google believes AI will be the most powerful engine fordiscovery that the web has ever seen.' If these AI tools deliver as promised, users may findthemselves spending less time searching—and more time simply receiving theanswers they need. In Google's future, your next deep dive into the web mightjust start and end with one query, handled entirely by AI.
The Verge
20-05-2025
- Business
- The Verge
Google is bringing an ‘Agent Mode' to the Gemini app
Google is bringing an 'Agent Mode' to the Gemini app and making some updates to its Project Mariner tool, CEO Sundar Pichai announced at Google I/O 2025. Project Mariner, Google's AI agent tool that can search the web for you, can now oversee up to 10 simultaneous tasks, Pichai said. Pichai also shared details on a feature called Teach and Repeat. 'This is where you can show your task once, and it learns a plan for similar tasks in the future.' Project Mariner will be available 'more broadly' this summer, he said. With Agent Mode in the Gemini app, you'll be able to give it a task to complete and the tool will go off and do it on your behalf, like with other AI agents. Pichai discussed an example of two people looking for an apartment in Austin, Texas. He said that the agent can find listings from sites like Zillow and use Project Mariner when needed to adjust specific filters. An 'experimental' version of Agent Mode will be 'coming soon' for subscribers, Pichai said.
