Info
Aktivitäten
1812 Follower:innen
-
Frida Almogabar Sanchez hat dies geteiltY que sigo sin creermela!!! Last week, I took a trip to the Google I/O event in Berlin. I felt like an influencer going into a brand trip... My inner geek was so thrilled to witness all of the latest updates and releases from the I/O! My absolute highlight of the day was my interview with my brilliant colleague Patrick Löber from Google DeepMind. I got to ask him my pressing questions around AI. Check out the full video below to see our interview and my recap of the event 👇 #GoogleDACH #GoogleIO #DeepMind #TechRecap #AItea #WomenInTech
-
Frida Almogabar Sanchez hat dies geteilt👀Frida Almogabar Sanchez hat dies geteiltWie ist es so, als Developer auf der 𝗚𝗼𝗼𝗴𝗹𝗲 𝗜/𝗢 𝗖𝗼𝗻𝗻𝗲𝗰𝘁 unterwegs zu sein?💥 AI Expertin Frida Almogabar Sanchez von Google Cloud hat es unserer Kollegin Pia D. in Berlin verraten. Hier kommt ihre Recap 👇 #GoogleIOConnect #AI #Developer #Google #BerlinGoogle I/O Connect Recap mit AI Expertin Frida Almogabar SanchezGoogle I/O Connect Recap mit AI Expertin Frida Almogabar Sanchez
-
Frida Almogabar Sanchez hat dies geteiltBerlin was hot, my laptop was hotter. 🔥 I really love to meet nerdy people like me who are all focused on one thing: creating things that actually matter. That was the vibe at Google I/O Connect today. It was so cool to meet so many people who just get it. We're all here to build and to solve problems. Berlin, you've been fabulous. Now to give my laptop (and my brain) a cool down. Stay tuned—a full recap video with all the good stuff is coming your way soon, also stay tuned with Google DACH as you might see me there 😜! #GoogleIO #IOConnect #Berlin #DevCommunity #AI #Builder #GoogleDACH
-
Frida Almogabar Sanchez hat dies geteiltLet's go Frankfurt! Qué locura de energía ayer! De verdad, otro nivel. 🔥 Stepping up for my keynote at the Google Cloud Summit DACH '26 and looking out at that room was incredible. Honestly, we spend so much time behind our screens tuning models and writing code, but nothing beats the spark that happens when this community gets together in person. The vibe yesterday was unmatched, and for my talk, I really wanted to bring something special to the stage. I showed a live demo of Antigravity—our autonomous coding agent. Showing what autonomous agents can actually do in real-time is always a thrill, and seeing the room's reaction to it was the absolute highlight of my day!! We aren't just talking about the future of engineering anymore; we are actively building it. Huge thanks to everyone who came out, brought their energy, and shared the day with us, especially Diana Nanova for sharing the stage with me! Hasta la próxima Google DACH! Keep your data clean, your models sharp, and your standards high. 💻✨ #CloudSummit #AIEngineering #GoogleCloud #AIAgents #FrankfurtTech #GoogleDACH
-
Frida Almogabar Sanchez hat dies geteiltOMG, I am incredibly thrilled & honored to announce that I will be taking the main stage as one of the keynote speakers at the Google DACH Cloud Summit in Frankfurt on June 9th! 🚀☁️ Join me as we dive into the future of technology. We will explore how the latest AI and cloud innovations are transforming the digital landscape, and I'll be sharing real-world insights and demos in my keynote session. I can't wait to connect with all of you, exchange ideas, and explore the cutting-edge of Generative AI and Cloud tech together! Check out the amazing speaker lineup and save your spot here: https://lnkd.in/dA-VQFH8 #GoogleCloudSummit #AI #CloudComputing #GoogleCloud #TechInnovation #Frankfurt #GoogleDACH
-
Frida Almogabar Sanchez hat dies geteiltThey say the definition of insanity is doing the same thing and expecting different results. But in AI engineering, it's just called 'prompting.' Isn't there a more structured way to build? We’ve all been there—overfitting that system prompt like a pair of shoes that are just a size too small, hoping it’ll handle every edge case. But maybe the secret to a lasting relationship with AI isn't more talk... it's more structure. 💅 Welcome to ADK 2.0! We're officially moving from the messy drama of prompt-based instructions to the chic, organized world of graph-based workflows. In my latest video below, I walk through a few short demos and break down my favorite new features from the release: ✨ Nodes & Edges: We're combining the raw reasoning of an LLM with the reliable, deterministic charm of plain Python code ✨ Code-First Approach: Everything is a Node. With Pydantic validation and smart parameter injection, transforming your code into a workflow is easier than switching from day-to-night attire. 👗 ✨ Collaboration API: Say hello to sophisticated multi-agent workflows where Coordinator agents delegate and specialized "leaf" agents deliver. No peer drama 👯♀️ Watch the video to see it all in action. ADK 2.0 gives you the safety net to build for the enterprise, ensuring your agents are dependable Because in AI, reliability never goes out of style. 🥂👇 and yes the cute little ADK(at) video was created by Omni - how cool is that? #AI #ADK2 #AgenticAI #GoogleCloud #GenerativeAI #WomenInTech
-
Frida Almogabar Sanchez hat dies geteiltIn tech, we love a flashy breakthrough. But as we hosted the VCI (Verband der Chemischen Industrie) yesterday in Berlin, the conversation was focused on something far more enduring: building AI systems that actually last in production. 🧪⚙️ Taking AI from a nice demo to a production-grade system in a heavy-duty sector like chemicals is a completely different kind of challenge. The vibe at the Google AI Center was focused entirely on security, sovereignty, and scale. During my keynote, we broke down what it actually takes to move beyond simple chatbots using the Gemini Enterprise Agent Platform: 1. Multi-Agent Topologies: Shifting from basic prompts and RAG to coordinated multi-agent setups 2. Centralized Control: Establishing zero-trust governance in production with Agent Identity and Agent Gateways 3. Observability & Traceability: Peeling back the curtain on complex agent reasoning loops. We talked about how to monitor, debug, and trace every single step to keep these autonomous systems predictable and stable (which sparked some of the most intense, technical questions of the day!). The "Secure AI Imperative" isn't a temporary trend; it’s the baseline. Big thanks to Ute Blauth for organizing and inviting me. Seasons change and trends come and go, but a robust architecture never goes out of style. Let’s be completely real—if your multi-agent orchestration is as chaotic as your group chat, we need to talk. #CodeAndCulture #GeminiEnterprise #AgenticAI #MultiAgentSystems #ChemicalIndustry #AIObservability #AISovereignty
-
Frida Almogabar Sanchez hat dies geteiltIs it just me, or is there nothing less attractive than a model that takes forever to reply? 💅 In tech, we’ve all been seduced by the biggest, most high-maintenance names on the market. But Google DeepMind just dropped Gemini 3.5 Flash, and suddenly, the heavy-duty flagships are looking a little too exhausting. Why waste precious compute waiting for a massive model to finish its internal monologue? 3.5 Flash is pure intelligence density—flagship-level reasoning packed into a sleek, high-velocity footprint that actually outperformed Gemini 3.1 Pro on core developer benchmarks. The details I'm completely obsessed with: - Built for agents: It handles complex, multi-step tool chains without losing its cool - High-fashion throughput: Run dense iteration cycles faster, cheaper, and with zero lag The pace of model iteration is wild right now, but this release proves that optimizing for speed doesn't mean compromising on deep reasoning🙂↕️🙂↕️ Are you testing 3.5 Flash in your dev workflows yet? How do you see it shifting your architecture? 👇Frida Almogabar Sanchez hat dies geteiltGemini 3.5 is our newest family of models combining frontier intelligence with real-world action. The first release is 3.5 Flash, our strongest model yet for agents and coding. It delivers fast, consistent performance that rivals other leading choices – at a fraction of the price. It can plan and reason across massive codebases and deploy subagents to work in parallel over a long horizon. It also outperforms 3.1 Pro on coding and agentic benchmarks. We’re rolling out 3.5 Flash to everyone in the Gemini app and AI Mode in Google Search. Developers can start building in Google Antigravity and via the Gemini API in Google AI Studio.
-
Frida Almogabar Sanchez hat dies geteiltI couldn’t help but wonder… in a city full of chatbots that talk your ear off, can we ever truly find an agent that’s ready to commit? Madre mia, que energía! We just wrapped our first stop of the Agent Factory 2026 series at Google Munich, and I feel so energized and inspired. I spent 48 hours watching developers and business leads try to navigate the pool of the tech world: Pilot Purgatory. We’ve all been there. You meet a chatbot, it’s charming, it says all the right things, but when it’s time to actually do something—like trigger a workflow or handle a complex API—it ghosts you. I watched squads move past the 'small talk' of simple prompts and get into the serious stuff—topology, memory banks, and the kind of grounding that keeps an enterprise from falling apart. Because let’s be honest: in 2026, if your agent doesn't have a backbone of solid architecture, it’s just a digital accessory you’ll eventually leave in the back of your closet. Seeing these teams turn 'what if' into 'production-ready' was better than anything I could have asked for. It was raw, it was technical, and it had that specific taste that only happens when you stop yapping and start building. Congrats to all of the winners and shout out to the Jochen Schweizer/my days team for getting the first prize! Also massive thanks for all of the people who make it possible Tanja C. Svajda, Alexander Erfurt, Sanchit Alekh, Oliver Lintner and soo many more. And special thanks to Wieland Holfelder for kicking us off! Munich, you were a perfect fit. 🇨🇭 Zurich, I’m packing my bags. The factory floor is waiting. #AgentFactory2026 #GoogleCloud #MunichTech #GenAI #AgenticSystems #CodeAndCulture #VertexAI
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt dasThe @Google Health API got a CLI! `ghealth` is an open-source CLI for your Fitbit Air. You can now pull 40 health metrics from the terminal or pipe them straight into your AI agent. Pull sleep breakdowns and heart rate data into context. Use `--detail` flag on sleep gives you per-stage data (awake, deep, REM), which is useful for spotting patterns week over week. ``` npx skills add Google-Health-API/google-health-cli ghealth setup ``` Setup takes a few minutes. Wizard handles GCP project, OAuth, scopes. Repository: https://lnkd.in/eyiFA3W7
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt dasAre we just using AI to do what we’ve always done, only faster? Or are we ready to explore the next phase of creative transformation? 💡 Just in time for #CannesLions2026, we’ve launched The Creativity Edition of Frontier CMO. We sat down with 20 creative leaders from across the industry to explore how they are moving beyond simple automation to help scale ambition, build connected creative systems, and support business growth. Download the full guide to read their insights 👇 https://goo.gle/4uLOY2J #GoogleBeach #GooglerInsights #GoogleEmployee #GoogleAI
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt dasBlog post #45. "#Agent2Agent ecosystem: #GeminiEnterprise Core Assistant as an #A2A Agent" As the virtual workforce expands, we need a robust central nervous system to manage enterprise productivity. #GeminiEnterprise is uniquely positioned to fill this gap by playing a dual role in the Agent-to-Agent (#A2A) ecosystem. On one hand, it acts as an intuitive gateway (A2A client) that connects external agents with company employees. On the other, its core assistant simultaneously serves as an A2A server—a sophisticated Enterprise #RAG agent that your custom, in-house applications can orchestrate and connect to. In this blog post, which is next in a series of my A2A Deep Dive Series, I explore how to practically integrate this architecture. The walkthrough covers the actual implementation details, showing how to: • Utilize the #AgentRegistry as the core governance and discovery layer for AI agents and #MCP servers on #GCP. • Browse and resolve A2A agents directly using the Python SDK. • Compose a custom multi-agent orchestrator that leverages the Gemini Enterprise Core Assistant as a knowledge specialist using the Google #AgentDevelopmentKit (#ADK). 👉 Link to blog post in comments. #A2A #GeminiEnterprise #GoogleCloud #AIArchitecture #AgentRegistry #ADK
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt dasHow do we separate reality from AI noise? The other day, I pranked my family group chat by sending them a picture of me skydiving. They almost believed it because the details and accuracy. In the end, the only thing that gave me away was the fact that I am absolutely terrified of heights! 🪂 Jokes aside, it made me stop and think. With image, sound and video models getting this accurate, distinguishing reality from fake content is becoming sometimes genuinely difficult and dangerous. It can feel overwhelming, but transparency tools are becoming crucial to helping us maintain our critical thinking and judgment online. This is one of the reasons why I've been following so interested the latest announcements for #SynthID. For those that don't know, SynthID is a digital watermarking technology that embeds invisible signals into AI content so you can easily trace its origin. Here are my 3 takeaways of the latest updates on SynthID: 1️⃣ My biggest highlight: major AI platforms like #OpenAI, #ElevenLabs, and #Kakao are now bringing SynthID technology into their own models to watermark their content. 2️⃣ You can now use daily tools like #Google Search, Circle to Search, and #Gemini in #Chrome to instantly ask if an image was made with AI. 3️⃣ Through global C2PA* standards, #Meta will soon automatically recognize and label original, unedited photos shot natively on Pixel phones when you share them on Instagram. *C2PA is the Coalition for Content Provenance and Authenticity. To put it in simple words, it's like a digital "nutrition label" or a verifiable chain-of-custody for photos, videos, audio or documents. Knowing where media comes from allows us to stay safe while keeping our analytical edge sharp. Seeing major industry players team up to build a safer digital ecosystem is exactly the kind of proactivity we need right now. Do you use tools like SynthID to verify digital content from your feed? Let me know your thoughts below. 👇 #ArtificialIntelligence #TechCommunity #DigitalTrust
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt das🏋️♀️ AI Training ≠ AI Adoption – warum Trainings allein nicht reichen Heute habe ich beim AI Camp (Nürnberg Digital Festival NUEDIGITAL) gemeinsam mit Sarah Fried vom BAMF einen Vortrag zu AI Adoption gehalten – ein Thema, das aktuell viele Organisationen und auch einige von euch, die heute dabei waren und tolle Diskussionen & Fragen gestellt haben, beschäftigt. Unsere Kernbotschaft heute: AI Adoption ist wie Fitness-Training. Ein Gym-Abo alleine (= Tool-Zugang) macht niemanden fit – und 1-2 KI-Trainings erzielen keine nachhaltige Verhaltensänderung. Es braucht eine ganzheitliche, menschenzentrierte Herangehensweise ➡️ heute in 6 "AI-Fitness Regeln" verpackt: 1️⃣ Access is prerequisite: ohne Zugang zu KI-Tools keine Adoption. Stattdessen "License to AI" für alle. 2️⃣ AI Training ≠ AI Adoption: Adoption ist eine langfristige Verhaltensänderung (Habit). Unser IBM AI Adoption Loop denkt AI Einführungen deshalb nicht nur holistisch und menschenzentriert, sondern vor allem iterativ. 3️⃣ Domain- & role-specific trainings: ein Trainingsplan für alle funktioniert im Sport nicht, bei KI auch nicht. Personaspezifische Lernpfade schaffen den echten Unterschied. 4️⃣ Stay safe, secure & compliant: klare Regeln für Daten, Zugriff und Modelle verhindern "Unfälle" und schaffen Vertrauen. 5️⃣ Train together: Social Learning, Communities und Change Agents machen KI-Lernen nachhaltiger – #continuity – und bringen auch mehr Spaß! 6️⃣ Measure progress & measure what matters: Nicht nur Lizenzaktivierungen, sondern Nutzung, Engagement und Impact messen. Ein großes Dankeschön geht an das Bundesamt für Migration und Flüchtlinge (BAMF) für die tolle Zusammenarbeit, an meine IBM Kolleg:innen Lisa Meingast, Tim Junge und Clara Störrle für die Einladung und die gemeinsame Zeit, und an das gesamte AI Camp / NUE Digital Festival Team für die großartige Organisation und Atmosphäre! 🙌 Denkt dran: Adoption happens by doing 💪. #TrainyourAImuscle #AIAdoption #ChangeManagement #KI #AICamp #NUEDigitalFestival #IBMConsulting #HumanFriendlyAutomation
-
Frida Almogabar Sanchez gefällt dasFrida Almogabar Sanchez gefällt dasThe last weeks have been so busy with exciting Google things that I missed my 7th Googleversary by one month, but better late than never 😀 And there it is, an excited Noogler 7 years ago not even close to having an idea what the next years will look like 🎉 For sure not expecting that they will be delivering #AI talks on the biggest stages wearing a shiny glitter jacket 😅 And yet, here we are doing exactly this and much more, loving and cherishing every day 💜 To many more exciting years at Google! If you love starting with a blank sheet of paper, solving complex technical puzzless for customers and like conquering the big stages (in your own style 🙃 ) our team is the perfect place to grow. 👉 Check out our open roles: https://lnkd.in/eWDhmtas #LifeatGoogle #Cloud #Hiring
Berufserfahrung und Ausbildung
-
Google
Gesamte Berufserfahrung von Frida Almogabar Sanchez anzeigen
Jobbezeichnung, Beschäftigungsdauer und mehr ansehen.
oder
Bescheinigungen und Zertifikate
Ehrenamt
-
-
Mentor
Kinderhelden
– 1 Jahr
Kinder
Serve as a mentor to and helping them with their path.
Kurse
-
Big Data & Analytics
-
-
IT Management
-
-
Software Management
-
Auszeichnungen/Preise
-
Major Outstanding Technical Achievement Award
IBM
The program celebrates exceptional technical contributions from Consulting teams who have created great client value and improved the IBM brand through their innovative solutions. These individuals and teams have applied their technical skills and insights to create and apply intellectual capital, reusable assets, and innovative techniques, creating higher and differentiated value and have explored scaling new technology for client success. As a result of these groundbreaking solutions, IBM…
The program celebrates exceptional technical contributions from Consulting teams who have created great client value and improved the IBM brand through their innovative solutions. These individuals and teams have applied their technical skills and insights to create and apply intellectual capital, reusable assets, and innovative techniques, creating higher and differentiated value and have explored scaling new technology for client success. As a result of these groundbreaking solutions, IBM Consulting has created additional sales, delivered superior client satisfaction, and increased market share. Notable contributions this year include Digital Twin technology, Artificial Intelligence, Large Language Models, IoT, cloud based serverless technology, event-driven architecture and virtual agents.
Sprachen
-
Spanisch
Muttersprache oder zweisprachig
-
English
Verhandlungssicher
-
Deutsch
Verhandlungssicher
Erhaltene Empfehlungen
2 Personen haben Frida Almogabar Sanchez empfohlen
Jetzt anmelden und ansehenWeitere ähnliche Profile
Weitere Beiträge entdecken
-
Eve H.
🚀 Real-time #analytics at 14.5 GB/s? We made it happen. Using Redpanda + Snowflake Streaming, we pushed 3.8 billion messages into a single Snowflake table with P99 latency under 8 seconds — all in less than a day of setup. Read the full benchmark to see how we achieved near #realtime analytics at massive scale. 👉 https://lnkd.in/dRptaBM6
-
Matt Aslett
New ISG Software Research Analyst Perspective: IBM Moves for Confluent to Add Event-Driven Context for AI As I previously stated, although most enterprises are reliant on batch data processing, it is an artificial construct driven by the historical limitations of computing capabilities to generate and process data at the same time without impacting performance. While real-time data processing has previously been seen as a niche requirement for low-latency applications, it is increasingly being adopted as a primary approach to data processing that enables enterprises to operate at the speed of business by acting on events as they happen. If artificial intelligence (AI) agents are to fulfill their potential and become core components of the next generation of enterprise software, then AI agents will also need to automate decisions as business events occur in real time, rather than as data related to those business events is processed. This is the context for the recent announcement that IBM has signed a definitive agreement to acquire streaming data specialist Confluent for approximately $11 billion in cash. https://lnkd.in/eAbbEWyt
-
Florian Bigelmaier
Two observations after attending Databricks Data + AI World Tour in beautiful SAP Garden in Munich today: Agent development is getting easier: Vendors like Databricks (and many more by the way - esp. the hyperscalers) are heavily investing in tools that reduce the complexity for developers and data scientists of building reliable agents. --> one example "Agent Learning from Human Feedback" (ALHF) caught my attention. It's reminding obviously of RLHF, but it's not about fine-tuning models though. It's rather a Databricks-selected term that wants to say: We have tools in place that allow to reflect on agent behavior based on minimal human input/feedback on agent input-output pairs and some other measurements as well as best practices. The output for ALHF would be recommendations and improvements on the application level of an agent - optimizing "behavior controls" like evals and adjusting the behavior of the agent by minimal user input. Noone builds the perfect agent with a zero shot prompt - so I'm very much looking forward how this will be picked up also by other players in the market. Messaging evolution: Over the time that I have observed Databricks, I realized a little shift from a strong emphasis of the "Open Source Data Lakehouse" toward positioning rather as a future-ready, open Data & AI platform - backed by recent improvements, acquisitions, and new developments like Unity Catalog, Lakebase, and Lakeflow. While that might be "just" messaging, it shows a lot how the platform is used and where Databricks might see future growth potential. #DataAI #Databricks Felix Mutzl Philipp Hirn Tanya Nitzsche Sonya Vargas (Ruzheva) Kayli Berlin
3 Kommentare