Data Reply FR

AWS re:Invent 2024 has been a blast! Our time in Las Vegas was packed with insights, innovation, and connections: - We delivered 4 engaging speeches with AWS, NetApp, and Intel, showcasing real-world solutions. - We held meaningful leadership meetings, planting seeds for future collaborations. - We explored a wave of in-depth Generative AI announcements, fueling new possibilities for our customers’ projects. The energy and innovation at re:Invent have been truly inspiring. Thank you to all partners, customers, and attendees who joined us there. https://lnkd.in/eMbPAQWa #AWSreInvent #GenerativeAI #CloudComputing #reInvent | Airwalk Reply | Concept Reply US | Comsysto Reply GmbH | Data Reply FR | Data Reply IT | Data Reply UK | Data Reply DE | Discovery Reply | Retail Reply | Riverland Reply | Sense Reply | Spike Reply DE | Storm Reply | Storm Reply DE | Storm Reply, Inc.

👑 Swami Sivasubramanian is officially the king of Las Vegas tonight after the 17 incredible announcements on Day 3 at AWS re:Invent 2024! Let’s dive into the innovations AWS dropped today: 🚀 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫 𝐀𝐈 𝐇𝐲𝐩𝐞𝐫𝐏𝐨𝐝 𝐅𝐥𝐞𝐱𝐢𝐛𝐥𝐞 𝐓𝐫𝐚𝐢𝐧𝐢𝐧𝐠 𝐏𝐥𝐚𝐧𝐬: Forget about manually provisioning compute power—this new feature distributes training workloads across accelerators, saving weeks of training time and optimizing your budget. Efficiency level: 🔥. 𝐇𝐲𝐩𝐞𝐫𝐏𝐨𝐝 𝐓𝐚𝐬𝐤 𝐆𝐨𝐯𝐞𝐫𝐧𝐚𝐧𝐜𝐞: Now you can track and manage generative AI tasks with ease, ensuring your resources are used optimally. Plus, it's up to 40% cheaper—goodbye unnecessary costs! 𝐀𝐈 𝐀𝐩𝐩𝐬 𝐟𝐫𝐨𝐦 𝐀𝐖𝐒 𝐏𝐚𝐫𝐭𝐧𝐞𝐫𝐬: SageMaker now allows you to discover, deploy, and use third-party machine learning and generative AI applications with full security and no infrastructure hassle. 🛠️ 🌐 𝐁𝐞𝐝𝐫𝐨𝐜𝐤 𝐍𝐞𝐰 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬 Poolside Malibu Poolside Point LUMA Ray 2 Stable Diffusion 3.5 𝐃𝐚𝐭𝐚 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐢𝐨𝐧 & 𝐌𝐮𝐥𝐭𝐢𝐦𝐨𝐝𝐚𝐥 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠: Automatically generate insights from multimodal content like documents, images, and audio, and enhance response accuracy by parsing both text and visuals. 𝐆𝐫𝐚𝐩𝐡𝐑𝐀𝐆 𝐒𝐮𝐩𝐩𝐨𝐫𝐭: Combine RAG and graphs to create even more accurate generative AI responses. 𝐒𝐭𝐫𝐮𝐜𝐭𝐮𝐫𝐞𝐝 𝐃𝐚𝐭𝐚 𝐑𝐞𝐭𝐫𝐢𝐞𝐯𝐚𝐥: Natural language queries now work with your data lakes and warehouses to accelerate AI development while unifying enterprise data sources. Amazon Bedrock 𝐈𝐧𝐭𝐞𝐥𝐥𝐢𝐠𝐞𝐧𝐭 𝐏𝐫𝐨𝐦𝐩𝐭 𝐑𝐨𝐮𝐭𝐢𝐧𝐠 optimizes response quality and cost by automatically selecting the best foundation model from a family based on prompt complexity. Amazon Bedrock 𝐏𝐫𝐨𝐦𝐩𝐭 𝐂𝐚𝐜𝐡𝐢𝐧𝐠 reduces costs and latency by storing frequently used context, ideal for applications like document Q&A and coding assistants that require repeated context access. 𝐁𝐞𝐝𝐫𝐨𝐜𝐤 𝐌𝐚𝐫𝐤𝐞𝐭𝐩𝐥𝐚𝐜𝐞: Over 100 foundation models available for developers through a unified catalog! 🔍 𝐀𝐦𝐚𝐳𝐨𝐧 𝐊𝐞𝐧𝐝𝐫𝐚 𝐆𝐞𝐧𝐀𝐈 𝐈𝐧𝐝𝐞𝐱 𝐀𝐝𝐯𝐚𝐧𝐜𝐞𝐝 𝐒𝐞𝐚𝐫𝐜𝐡: Boost search accuracy and integrate across AWS generative AI services with 43 data sources supported. Search smarter, not harder! 🔒 𝐁𝐞𝐝𝐫𝐨𝐜𝐤 𝐆𝐮𝐚𝐫𝐝𝐫𝐚𝐢𝐥𝐬 𝐌𝐮𝐥𝐭𝐢𝐦𝐨𝐝𝐚𝐥 𝐓𝐨𝐱𝐢𝐜𝐢𝐭𝐲 𝐃𝐞𝐭𝐞𝐜𝐭𝐢𝐨𝐧: Now you can filter undesirable visuals and texts in your generative AI apps, ensuring responsible AI policies with configurable thresholds. 💡 𝐀𝐦𝐚𝐳𝐨𝐧 𝐐 𝐐 𝐃𝐞𝐯𝐞𝐥𝐨𝐩𝐞𝐫 𝐢𝐧 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫 𝐂𝐚𝐧𝐯𝐚𝐬: Build machine learning models using natural language. Just tell it what you need, and Q does the rest. 🧑💻 𝐐 𝐢𝐧 𝐐𝐮𝐢𝐜𝐤𝐒𝐢𝐠𝐡𝐭: Run complex business scenario analyses and get insights 10x faster than spreadsheets—talk about next-level business intelligence! At Data Reply FR, we’re thrilled to help our customers integrate these game-changing innovations.

🔥 Day 2 at AWS re:Invent 2024 – If Day 1 had you saying, “Wow, that’s cool,” Day 2 will have you shouting, “Take my money!” From turbo-charged AI chips to multi-agent orchestration that sounds like a spy movie, AWS is serving up innovations faster than I can say “latency reduction.” Buckle up for the highlights: 🌟 𝐀𝐦𝐚𝐳𝐨𝐧 𝐁𝐞𝐝𝐫𝐨𝐜𝐤 𝐌𝐨𝐝𝐞𝐥 𝐃𝐢𝐬𝐭𝐢𝐥𝐥𝐚𝐭𝐢𝐨𝐧: Speed up by 500% and cut costs by 75%, without sacrificing accuracy. Faster, smarter, cheaper—pick all three! 𝐀𝐮𝐭𝐨𝐦𝐚𝐭𝐞𝐝 𝐑𝐞𝐚𝐬𝐨𝐧𝐢𝐧𝐠 𝐂𝐡𝐞𝐜𝐤𝐬: Enhanced safety with responsible AI checks to build trust into AI applications 𝐌𝐮𝐥𝐭𝐢-𝐀𝐠𝐞𝐧𝐭 𝐂𝐨𝐥𝐥𝐚𝐛𝐨𝐫𝐚𝐭𝐢𝐨𝐧: Orchestrate multiple agents seamlessly for advanced workflows 🎨 𝐅𝐨𝐮𝐧𝐝𝐚𝐭𝐢𝐨𝐧 𝐌𝐨𝐝𝐞𝐥𝐬 6 𝐍𝐞𝐰 𝐀𝐦𝐚𝐳𝐨𝐧 𝐍𝐨𝐯𝐚 𝐌𝐨𝐝𝐞𝐥𝐬: Including Nova Canvas for images and Nova Reel for videos. These models: Match or exceed competitor performance Slash costs by 75% compared to market leaders Support 200+ languages (while others stay English-focused) Already powering 1,000+ Amazon applications Coming Soon: Speech-to-speech and any-to-any models by 2025. The future looks WOW! 😄 💻 𝐀𝐈 𝐂𝐡𝐢𝐩𝐬 𝐓𝐫𝐚𝐢𝐧𝐢𝐮𝐦2: Delivers 4x performance gains and cuts costs by 50% compared to Nvidia Apple joins as a major customer. 🍎 Deep collaboration with Annapurna Labs fuels innovation. Next-gen Trainium3 already in the works! 🛠️ 𝐀𝐦𝐚𝐳𝐨𝐧 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫 𝐍𝐞𝐱𝐭-𝐆𝐞𝐧 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫 𝐒𝐭𝐮𝐝𝐢𝐨: Unified tools for data engineering, analytics, and generative AI in one streamlined workspace 𝐒𝐚𝐠𝐞𝐌𝐚𝐤𝐞𝐫 𝐋𝐚𝐤𝐞𝐡𝐨𝐮𝐬𝐞: Simplifies analytics by integrating S3 data lakes with Redshift warehouses using open Apache Iceberg APIs and fine-grained access controls 🤖 𝐀𝐦𝐚𝐳𝐨𝐧 𝐐 Developer Updates: .NET porting transformations GitLab Duo integration Automated code reviews and unit test generation Documentation generation Business Updates: New workflow automation with 50+ action integrations Insights directly from databases and warehouses. 📊 𝐒3 𝐓𝐚𝐛𝐥𝐞𝐬 & 𝐃𝐚𝐭𝐚𝐛𝐚𝐬𝐞𝐬 𝐒3 𝐓𝐚𝐛𝐥𝐞𝐬: Fully managed Apache Iceberg tables for analytics workloads. Delivering 3x faster queries and 10x higher transactions/second 𝐃𝐲𝐧𝐚𝐦𝐨𝐃𝐁 𝐆𝐥𝐨𝐛𝐚𝐥 𝐓𝐚𝐛𝐥𝐞𝐬: Multi-region strong consistency now in preview  𝐀𝐮𝐫𝐨𝐫𝐚 𝐃𝐒𝐐𝐋: New serverless, distributed SQL database with active-active high availability 🔥 AWS is delivering transformative tools to power the future of AI, analytics, and cloud computing—and at Data Reply FR, we’re here to help you integrate these groundbreaking features into your business. From faster AI chips to unified analytics platforms, we work with our customers to turn innovation into action. Which of these announcements are you most excited about? Let’s discuss! PS: The image you see here? It’s not just any picture—it was generated using the NOVA Canvas model on Bedrock. #AWSreInvent #AIInnovation #CloudComputing #MachineLearning #GenerativeAI

Just finished attending Matt's keynote at #AWS #reInvent2024, and I’m blown away🤯 by the incredible innovations announced! Here are some highlights I’m particularly excited about: 🚀𝗔𝗺𝗮𝘇𝗼𝗻 𝗕𝗲𝗱𝗿𝗼𝗰𝗸 𝗠𝘂𝗹𝘁𝗶-𝗔𝗴𝗲𝗻𝘁 𝗖𝗼𝗹𝗹𝗮𝗯𝗼𝗿𝗮𝘁𝗶𝗼𝗻: This takes productivity to a whole new level by enabling teams of agents to collaborate seamlessly on complex, multi-step tasks. Building, deploying, and orchestrating agents has never been easier or faster. 🤖𝗔𝘂𝘁𝗼𝗺𝗮𝘁𝗲𝗱 𝗥𝗲𝗮𝘀𝗼𝗻𝗶𝗻𝗴: A powerful new mechanism for ensuring systems perform as intended. It’s fascinating to see AI being used to not only automate but also verify outcomes, making systems smarter and more reliable. 💡𝗔𝗺𝗮𝘇𝗼𝗻 𝗡𝗼𝘃𝗮 𝗠𝗼𝗱𝗲𝗹𝘀 : A versatile lineup of foundation models to address diverse business needs. 🔥𝗔𝗺𝗮𝘇𝗼𝗻 𝗕𝗲𝗱𝗿𝗼𝗰𝗸 𝗠𝗼𝗱𝗲𝗹 𝗗𝗶𝘀𝘁𝗶𝗹𝗹𝗮𝘁𝗶𝗼𝗻 : A game-changer for model efficiency, enabling knowledge transfer from large to smaller models. Models can now run up to 500% faster while being 75% more cost-effective. This is a huge leap forward for scalability! 📊𝗔𝗺𝗮𝘇𝗼𝗻 𝗦𝟯 𝗧𝗮𝗯𝗹𝗲𝘀: Introducing a new bucket type designed specifically for Apache Iceberg tables, offering 3x faster query performance and up to 10x higher transactions per second. This is a major leap forward for analytical workloads, making data access faster and more scalable than ever before. 🔀 I was also deeply impressed by Matt’s thoughts on "𝘁𝗵𝗲 𝘁𝘆𝗿𝗮𝗻𝗻𝘆 𝗼𝗳 𝗢𝗥" VS "𝘁𝗵𝗲 𝗴𝗲𝗻𝗶𝘂𝘀 𝗼𝗳 𝗔𝗡𝗗". #AWS aim to embrace both, innovating in ways that allow us customers to deliver comprehensive solutions without having to choose one over the other. This mindset drives the incredible innovation and value AWS consistently provides. AWS continues to push boundaries, delivering tools that make it easier to build, scale, and innovate and #Reply is excited to see how these new capabilities will transform workflows in AI, data management, and analytics and drive business value for our customers. Let’s connect if you’d like to dive deeper into these exciting innovations! 🚀🌟

Day 1 of AWS re:Invent in Las Vegas is over, and it's been as electrifying as a Vegas slot machine win! 🎰 But instead of coins, we're hitting jackpots in Machine Learning and Generative AI. Here’s a quick roundup of the biggest announcements: 🤖 𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗶𝘃𝗲 𝗔𝗜 𝗼𝗻 𝗕𝗲𝗱𝗿𝗼𝗰𝗸: 𝟭/𝗥𝗔𝗚 𝗚𝗲𝘁𝘀 𝗝𝘂𝗱𝗴𝘆: Bedrock introduced RAG evaluation with both labeled and unlabeled test data, and—drumroll—LLM as a judge! This feature leverages an LLM to evaluate responses, outperforming statistical approaches on consistency with human grading: https://lnkd.in/eRmBFW49 𝟮/ 𝗠𝗼𝗱𝗲𝗹-𝗚𝗲𝗻𝗲𝗿𝗮𝘁𝗲𝗱 𝗙𝗶𝗹𝘁𝗲𝗿𝘀: Your queries just got smarter. LLMs can now transform user queries into filters and questions. Translation: it’s like having a genius travel buddy—"Show me pet-friendly hotels in Paris" translates to "pet_friendly=True" and a query for Paris hotels. Voilà! https://lnkd.in/eS5WVHNz 𝟯/ 𝗠𝗲𝘁𝗮𝗱𝗮𝘁𝗮 𝗘𝗺𝗯𝗲𝗱𝗱𝗶𝗻𝗴: Choose your metadata, embed it, and unlock powerful query capabilities across text and metadata. Think "next-level search engine" vibes: https://lnkd.in/egF9k6UC 𝟰/𝗥𝗲𝗿𝗮𝗻𝗸𝗶𝗻𝗴: Faster, better responses? Yes, please. Reranking improves latency and response quality: https://lnkd.in/eJUZxWJb 𝟱/ 𝗖𝘂𝘀𝘁𝗼𝗺 𝗖𝗼𝗻𝗻𝗲𝗰𝘁𝗼𝗿𝘀: Bring data from anywhere streams and APIs, and you name it—into Knowledge Bases. Because diverse data deserves diverse solutions: https://lnkd.in/evCb33fs  𝟲/ 𝗦𝘁𝗿𝗲𝗮𝗺𝗶𝗻𝗴: Life-like interactions are here! Response streaming reduces time-to-token, giving apps more real-time appeal: https://lnkd.in/ez3dCGqk ⚙️ 𝗢𝗻 𝗦𝗮𝗴𝗲𝗠𝗮𝗸𝗲𝗿: The new “Scale Down to Zero” feature lets SageMaker Inference Endpoints scale to zero when not in use, saving costs while staying ready to scale up. Practical and efficient: https://lnkd.in/etdwEWrp 💡 𝗔𝗜 𝗖𝗵𝗶𝗽𝘀: Say hello to 𝗧𝗿𝗮𝗶𝗻𝗶𝘂𝗺𝟮, now generally available! With 2.6 petaflops of compute and 96 GB of high-speed memory, it’s like handing your AI a sports car. 🚗💨: https://lnkd.in/ewBu5bzZ Exciting times ahead for Machine Learning and Generative AI! Can’t wait to see how you’ll leverage these updates at Data Reply FR #AWS #reInvent2024 #ML #GenerativeAI #Bedrock #SageMaker #AIChips #TechInnovation

Here we go 🤠—kicking off an exciting week at #AWS #reInvent in Las Vegas! Can’t wait to dive into the latest cloud innovations, connect with brilliant minds, and hear what’s next in generative AI. Vegas is buzzing, so What happens in Vegas stays in Vegas 👌…unless it’s a groundbreaking AWS announcement. Let’s make it a great week!🚀⭐️

Go go go Data Reply.

Here we go 🤠—kicking off an exciting week at #AWS #reInvent in Las Vegas! Can’t wait to dive into the latest cloud innovations, connect with brilliant minds, and hear what’s next in generative AI. Vegas is buzzing, so What happens in Vegas stays in Vegas 👌…unless it’s a groundbreaking AWS announcement. Let’s make it a great week!🚀⭐️

The wait is over—Data Reply FR is ready to make the most of AWS re:Invent at Las Vegas! 🚀 We’re excited to connect, collaborate, and explore the latest in cloud innovation at this incredible event. From inspiring keynotes to hands-on workshops, the opportunities to learn and grow are endless. Got your badges ready? We sure do! See you there folks Hedi Aloui Amine AIT EL HARRAJ Wael Saideni #AWSreInvent #DataReply #CloudInnovation #Networking #ReadyToLearn #LasVegas

🎉 Amazon Web Services (AWS) GEN AI DAY : Retour sur un événement inspirant dédié aux startups et à l’innovation à STATION F ! Merci à toutes les startups, partenaires, et innovateurs qui ont rejoint Data Reply FR et Amazon Web Services (AWS) pour cet événement unique. 🚀 Au programme : ✨ Des discussions passionnantes sur la co-construction de solutions innovantes. ⚡ Des échanges sur l'accélération des roadmaps grâce à la synergie entre Data Reply FR et Amazon Web Services (AWS). 🌐 Une communauté engagée pour transformer les défis en opportunités. Et puis, l'agilité des startups n'est plus à discuter. Nous avons réussi le défi de nous engager avec des startups de 1er niveau durant l'événement et nous allons démarrer des nouvelles aventures d'innovation dans quelques jours seulement. Nous sommes impatients. Un grand merci à tous les participants et partenaires pour cette expérience enrichissante. 🙏 Edouard MARTIN Hadrien Robino Julien DHAINAUT Nicolas Pfeiffer Thierry Pierre Joel Farvault Vivien de Saint Pern Thomas Caquineau Thomas Villareal 📩 Si vous souhaitez en savoir plus ou explorer comment nous pouvons collaborer, n’hésitez pas à nous contacter ! Hedi Aloui Tahiti Blondin Ahmed Cheikh Rouhou Maëlle T. Alessandro Danesi Amine AIT EL HARRAJ #Innovation #Startups #AWS #DataReply #Collaboration #Succès