<?xml version="1.0" encoding="UTF-8"?><rss xmlns:dc="http://purl.org/dc/elements/1.1/" xmlns:content="http://purl.org/rss/1.0/modules/content/" xmlns:atom="http://www.w3.org/2005/Atom" version="2.0" xmlns:cc="http://cyber.law.harvard.edu/rss/creativeCommonsRssModule.html">
    <channel>
        <title><![CDATA[Stories by Fabio Domingues on Medium]]></title>
        <description><![CDATA[Stories by Fabio Domingues on Medium]]></description>
        <link>https://medium.com/@fbiodomingues?source=rss-d235cb83dbd9------2</link>
        <image>
            <url>https://cdn-images-1.medium.com/fit/c/150/150/2*BXl6Q12cnY3vovx1fftH_w.jpeg</url>
            <title>Stories by Fabio Domingues on Medium</title>
            <link>https://medium.com/@fbiodomingues?source=rss-d235cb83dbd9------2</link>
        </image>
        <generator>Medium</generator>
        <lastBuildDate>Sat, 21 Nov 2026 16:07:01 GMT</lastBuildDate>
        <atom:link href="https://medium.com/@fbiodomingues/feed" rel="self" type="application/rss+xml"/>
        <webMaster><![CDATA[yourfriends@medium.com]]></webMaster>
        <atom:link href="http://medium.superfeedr.com/" rel="hub"/>
        <item>
            <title><![CDATA[Saving Your Mental Energy: Why Chat Wins]]></title>
            <link>https://medium.com/@fbiodomingues/saving-your-mental-energy-why-chat-wins-754d6a7dbd4e?source=rss-d235cb83dbd9------2</link>
            <guid isPermaLink="false">https://medium.com/p/754d6a7dbd4e</guid>
            <dc:creator><![CDATA[Fabio Domingues]]></dc:creator>
            <pubDate>Mon, 12 May 2025 10:56:03 GMT</pubDate>
            <atom:updated>2025-05-12T10:56:03.214Z</atom:updated>
            <content:encoded><![CDATA[<p>The rise of artificial intelligence (AI) has brought us to an essential question: how will we interact with AI effectively in the future? Initially, many viewed conversational interfaces like ChatGPT merely as prototypes — early steps towards integrating AI into existing apps and websites. However, I argue that, contrary to this perception, the chat interface (through text, voice, video, and potentially other emerging modalities) will emerge as our primary and preferred way to interact with AI systems.</p><p>Many believe the next step in AI interfaces will be tightly integrating AI directly into each application, enhancing context-specific tasks. However, I propose a different vision: the future belongs to chat. Rather than fragmented experiences, we’ll increasingly rely on unified, conversational interfaces that adapt fluidly across all interactions.</p><h3>Energy vs. Time: The Core Advantage of Chat</h3><p>Why is the conversational interface becoming preferable, even when it might take longer to perform tasks?</p><p>The key lies in cognitive energy. Although using chat interfaces might initially seem slower, the mental effort required is significantly lower. Simply put, chatting is natural and intuitive for humans. It reduces cognitive load, eliminating the need to constantly learn, adapt, or switch contexts between multiple graphical user interfaces.</p><p>Steve Krug’s book <em>“Don’t Make Me Think”</em> emphasizes this principle clearly: the best interfaces are those that minimize the user’s cognitive effort. For example, Krug discusses how even small points of friction, like unclear navigation or complex layouts, can disrupt the user experience. The chat interface exemplifies this concept by requiring minimal thought about how to use it, providing a consistent and straightforward interaction model.</p><p><strong>Imagine you’re relaxing on your couch, absorbed in a book or watching your favorite series. You’re thirsty, but getting up means disrupting your comfort and focus. Your friend is already standing nearby, so naturally, you ask them to pass you a glass of water — even if you need to wait a moment longer. This minor convenience illustrates why chat interactions appeal to us: they let us stay comfortably focused, avoiding the mental disruption of switching tasks, even if it takes slightly more time.</strong></p><p>This illustrates perfectly why chat interactions are appealing. Users exchange additional waiting time for reduced mental effort, making the interaction feel easier overall.</p><h3>The iPhone Revolution</h3><p>When Steve Jobs introduced the iPhone, he transformed how we think about interfaces. The touchscreen was revolutionary because it adapted dynamically to whatever task the user needed. Unlike fixed keyboards of earlier smartphones, the touchscreen provided the optimal interface for each scenario.</p><p>Similarly, the chat interface dynamically adapts to user interactions, offering flexibility and simplicity simultaneously. This adaptability minimizes mental energy, creating a consistent and comfortable experience.</p><h3>Switching Between Apps is Draining our Energy</h3><p>Opening multiple apps throughout the day — Gmail, Slack, calendars, task managers — requires cognitive energy. Each app has its own logic, layout, colors, and navigation. Although we’ve automated much of this mentally, the subconscious cognitive load remains significant.</p><p>Conversational AI assistants dramatically reduce this load. Personally, I’ve found myself increasingly using my AI assistant (Gizmo via Telegram) for tasks like managing emails or scheduling. Even though it’s technically slower than using dedicated apps, the mental effort required is significantly lower. Interestingly, this makes me feel less exhausted, enhancing overall productivity.</p><h3>Predicting the Future: Chat as the Dominant Interface</h3><p>Looking forward, interactions with AI will converge towards unified chat interfaces — personal assistants we become familiar and comfortable with. These assistants will communicate with various AI services (likely MCPs) on our behalf, streamlining our interactions and significantly reducing cognitive overhead.</p><p>As chat interfaces become central to our digital lives, an important question arises: Who will create the most dominant chat app? Could it be ChatGPT evolving into a comprehensive platform, WhatsApp leveraging its enormous global user base, or perhaps another contender entirely? Looking at China’s WeChat, which already integrates extensive AI-driven functionalities and services, offers a glimpse into how powerful and integral a unified chat interface could become globally. The race to build the definitive chat platform is just beginning.</p><h3>Conclusion</h3><p>The chat interface isn’t just another design option — it represents a fundamental shift in human-computer interaction driven by cognitive ease rather than raw efficiency. Companies, designers, and developers must begin rethinking their approaches, prioritizing mental energy conservation over mere speed or click efficiency.</p><p>We’re at the brink of an interface revolution where conversation, simple and natural, becomes the ultimate interaction model.</p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=754d6a7dbd4e" width="1" height="1" alt="">]]></content:encoded>
        </item>
        <item>
            <title><![CDATA[MCP: The Missing Piece That Finally Makes Your AI Useful]]></title>
            <link>https://medium.com/@fbiodomingues/mcp-the-missing-piece-that-finally-makes-your-ai-useful-2e4f20a5e012?source=rss-d235cb83dbd9------2</link>
            <guid isPermaLink="false">https://medium.com/p/2e4f20a5e012</guid>
            <category><![CDATA[agents]]></category>
            <category><![CDATA[ai]]></category>
            <category><![CDATA[llm]]></category>
            <category><![CDATA[mcp-server]]></category>
            <dc:creator><![CDATA[Fabio Domingues]]></dc:creator>
            <pubDate>Mon, 14 Apr 2025 14:04:08 GMT</pubDate>
            <atom:updated>2025-04-14T14:04:08.491Z</atom:updated>
            <content:encoded><![CDATA[<p><strong>Picture this</strong>: You wake up, and your virtual assistant has already scheduled your day, answered your emails, booked your flights, and even pointed out exactly where your latest project might need a few tweaks — without you having to lift a finger. It sounds futuristic, almost magical. But this isn’t sci-fi; it’s quietly becoming reality.</p><p>Behind this seamless, magical future lies an invisible revolution: a groundbreaking idea known simply as the Model Context Protocol.</p><figure><img alt="" src="https://cdn-images-1.medium.com/max/1024/1*RVNYU-ExMcJXE9FPRrQuig.png" /></figure><p>Never heard of it? You’re not alone. But soon, everyone might be talking about it. Imagine a universal translator, something that allows your AI to communicate instantly and clearly with every service you use — from GitHub to Google Calendar, from Slack to Spotify — without the usual headaches of setup, integrations, or technical jargon. This is exactly what the Model Context Protocol promises.</p><p>Right now, if you peek behind the curtain of your apps, you’ll find a tangled mess. Each service speaks a unique and complicated language. Integrating them takes enormous effort and resources, often resulting in frustration and confusion. What should be simple interactions become costly, complex, and painfully slow.</p><p>But imagine a world where integrations disappear into the background, where AI effortlessly navigates these interactions, making your life simpler, smoother, and infinitely more productive. That’s precisely what this new approach offers. It’s not just an upgrade; it’s a complete redefinition of how technology can and should serve us.</p><p>Yet, the adoption has been slow — hidden behind layers of technical complexity that make it difficult for everyday users and even developers to dive in. It’s akin to needing a complex home-built router just to use Wi-Fi; the barrier is high, but the potential payoff is enormous.</p><p>In my own experiments converting typical APIs into this new format, I uncovered something crucial: many of today’s digital tools aren’t even user-friendly enough for AI. They’re overly complex, opaque, and built without clarity in mind. For AI to truly serve us better, we need clearer, more intuitive interfaces. Ironically, making machines smarter begins with making our technology simpler and more human-friendly.</p><p>Tech giants already sense this shift. Google’s recent “Agent to Agent” initiative shows they’re betting big on this vision — a future where AIs communicate seamlessly not just with tools but also with each other, making our digital lives almost effortlessly efficient.</p><p>But why should this matter to you? Because soon, the apps and services you rely on daily will be fundamentally transformed by this invisible protocol. Your interactions will become effortless, your digital tasks more intuitive, and your overall experience remarkably smoother. Like Wi-Fi before it, this new standard will subtly yet profoundly reshape our relationship with technology.</p><p>So pay attention: the revolution isn’t loud or flashy — but it’s here, quietly reshaping your future, one seamless interaction at a time.</p><p>Here are two bold predictions about MCP:</p><ol><li><strong>MCP will replace APIs as the everyday tech term.</strong> Just as “Wi-Fi” replaced complicated technical terms for wireless connections, “MCP” could become the go-to word for how AIs connect to tools. You won’t just say, “Our software has an API,” you’ll say, “Our software supports MCP.”</li><li><strong>APIs will become AI-friendly by necessity.</strong> MCP will push companies to design tools specifically understandable by AI models. Imagine APIs becoming intuitive and human-like, making them not only AI-friendly but user-friendly too. MCP essentially becomes the “face” of the tools, turning software interactions into simple, conversational exchanges.</li></ol><figure><img alt="" src="https://cdn-images-1.medium.com/max/200/1*7Y-ZCTjZa2V-5V8nC8Fr3A.png" /><figcaption><a href="https://github.com/modelcontextprotocol">https://github.com/modelcontextprotocol</a></figcaption></figure><p>MCP: <a href="https://github.com/modelcontextprotocol">https://github.com/modelcontextprotocol</a></p><p>Announcing the Agent2Agent Protocol (A2A): <a href="https://developers.googleblog.com/en/a2a-a-new-era-of-agent-interoperability/">https://developers.googleblog.com/en/a2a-a-new-era-of-agent-interoperability/</a></p><img src="https://medium.com/_/stat?event=post.clientViewed&referrerSource=full_rss&postId=2e4f20a5e012" width="1" height="1" alt="">]]></content:encoded>
        </item>
    </channel>
</rss>