Altabel Group

A short roundup of AI news from this month. 𝗔𝗜 𝗶𝘀 𝗶𝗻𝗰𝗿𝗲𝗮𝘀𝗶𝗻𝗴 𝗱𝗲𝗺𝗮𝗻𝗱 𝗳𝗼𝗿 𝗰𝗹𝗲𝗮𝗻 𝗲𝗻𝗲𝗿𝗴𝘆 According to Axios, companies signed record volumes of clean energy deals in early 2026. One of the main reasons for this is the rapid growth of AI and data centres, which require huge amounts of electricity to run. AI is now affecting not only software and apps, but also energy infrastructure. 𝗦𝗽𝗼𝘁𝗶𝗳𝘆 𝘀𝘁𝗮𝗿𝘁𝗲𝗱 𝗮𝗱𝗱𝗶𝗻𝗴 𝗯𝗮𝗱𝗴𝗲𝘀 𝗳𝗼𝗿 ‘𝗿𝗲𝗮𝗹’ 𝗮𝗿𝘁𝗶𝘀𝘁𝘀 The platform is introducing verified artist badges so users can better distinguish human musicians from AI-generated music profiles. The interesting part is not the badge itself, but the reason behind it. AI music has become so widespread that streaming platforms now need new ways to indicate what is human-made and what is not. 𝗝𝗮𝗽𝗮𝗻 𝗔𝗶𝗿𝗹𝗶𝗻𝗲𝘀 𝗶𝘀 𝘁𝗲𝘀𝘁𝗶𝗻𝗴 𝗵𝘂𝗺𝗮𝗻𝗼𝗶𝗱 𝗿𝗼𝗯𝗼𝘁𝘀 𝗮𝘁 𝘁𝗵𝗲 𝗮𝗶𝗿𝗽𝗼𝗿𝘁 The airline started testing robots for moving baggage and cargo at Haneda Airport. This is a good example of robotics becoming part of everyday operations, especially in places dealing with staff shortages and growing passenger traffic. 𝗧𝗲𝗰𝗵 𝗴𝗶𝗮𝗻𝘁𝘀 𝗯𝗮𝗰𝗸 𝗻𝗲𝘄 𝗱𝗮𝘁𝗮 𝗰𝗲𝗻𝘁𝗿𝗲 𝗰𝗹𝗶𝗺𝗮𝘁𝗲 𝗶𝗻𝗶𝘁𝗶𝗮𝘁𝗶𝘃𝗲 Microsoft, Google, Amazon and Meta backed a new initiative that will use data centres as test beds for climate-friendly technologies. Elemental Impact plans to invest from $500,000 to $5 million in up to ten startups by 2027. This demonstrates that AI infrastructure is becoming an issue not just for IT, but for energy and climate too.

If you need proven specialists who can join active development, strengthen your team and contribute without a long onboarding period, let’s discuss!

Our colleague Dmitriy Denchikov shared a good example of AI making the distance between idea and visual concept much shorter. You no longer need to be an illustrator to show what you have in mind. Simple, but impressive.

𝗔 𝗻𝗲𝘄 𝗿𝗲𝘃𝗶𝗲𝘄 𝗼𝗻 𝗖𝗹𝘂𝘁𝗰𝗵 𝗳𝗿𝗼𝗺 𝗮 𝘀𝗼𝗳𝘁𝘄𝗮𝗿𝗲 𝗰𝗼𝗺𝗽𝗮𝗻𝘆 𝗱𝗲𝘃𝗲𝗹𝗼𝗽𝗶𝗻𝗴 𝗮𝗻 𝗲𝗻𝘁𝗲𝗿𝗽𝗿𝗶𝘀𝗲 𝗦𝗮𝗮𝗦 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺. The client specializes in software for governance, risk and compliance management. Their platform helps legal and compliance teams organize contracts, obligations, approvals, documentation and reporting in one structured digital environment. Altabel was responsible for building the DevOps foundation from scratch. In practical terms, we helped to make releases easier to manage: - separate environments were set up for development, testing and production; - updates became easier to prepare, check and release; - sensitive data and access rights were organized in a more controlled way. This review is valuable to us because it highlights something we consider essential in this type of cooperation: the work must be technically solid, but it must also appear clear, predictable and well coordinated to the client's team. 𝗙𝘂𝗹𝗹 𝗰𝗹𝗶𝗲𝗻𝘁 𝗿𝗲𝘃𝗶𝗲𝘄: https://lnkd.in/dA-4Mfpj

𝗢𝘂𝘁𝗮𝗴𝗲𝘀 𝗱𝗼𝗻’𝘁 𝘀𝘁𝗮𝗿𝘁 𝗮𝘁 𝗻𝗶𝗴𝗵𝘁. 𝗧𝗵𝗲𝘆 𝘀𝘁𝗮𝗿𝘁 𝗮𝘁 𝘁𝗵𝗲 𝗺𝗼𝗺𝗲𝗻𝘁 𝗼𝗳 𝗰𝗵𝗮𝗻𝗴𝗲. Yegor is a full-stack engineer with experience of migrating data models, integrating external services and updating core backend logic in e-commerce and fintech systems, often while those systems were live and handling active users and transactions. From Yegor’s experience: ‘𝘐𝘯 𝘮𝘰𝘴𝘵 𝘰𝘶𝘵𝘢𝘨𝘦𝘴 𝘐’𝘷𝘦 𝘴𝘦𝘦𝘯, 𝘵𝘩𝘦 𝘪𝘴𝘴𝘶𝘦 𝘸𝘢𝘴𝘯’𝘵 𝘵𝘩𝘦 𝘥𝘦𝘱𝘭𝘰𝘺𝘮𝘦𝘯𝘵 𝘪𝘵𝘴𝘦𝘭𝘧. 𝘐𝘵 𝘸𝘢𝘴 𝘩𝘰𝘸 𝘵𝘩𝘦 𝘤𝘩𝘢𝘯𝘨𝘦𝘴 𝘸𝘦𝘳𝘦 𝘪𝘯𝘵𝘳𝘰𝘥𝘶𝘤𝘦𝘥 𝘪𝘯𝘵𝘰 𝘵𝘩𝘦 𝘴𝘺𝘴𝘵𝘦𝘮. 𝘛𝘦𝘢𝘮𝘴 𝘴𝘵𝘪𝘭𝘭 𝘳𝘦𝘭𝘺 𝘰𝘯 ‘𝘣𝘪𝘨 𝘣𝘢𝘯𝘨’ 𝘶𝘱𝘥𝘢𝘵𝘦𝘴, 𝘸𝘩𝘦𝘳𝘦 𝘴𝘤𝘩𝘦𝘮𝘢𝘴, 𝘭𝘰𝘨𝘪𝘤 𝘰𝘳 𝘥𝘢𝘵𝘢 𝘮𝘰𝘥𝘦𝘭𝘴 𝘢𝘳𝘦 𝘤𝘩𝘢𝘯𝘨𝘦𝘥 𝘪𝘯 𝘢 𝘴𝘪𝘯𝘨𝘭𝘦 𝘴𝘵𝘦𝘱, 𝘢𝘯𝘥 𝘵𝘩𝘪𝘴 𝘪𝘴 𝘸𝘩𝘦𝘳𝘦 𝘴𝘺𝘴𝘵𝘦𝘮𝘴 𝘵𝘦𝘯𝘥 𝘵𝘰 𝘧𝘢𝘪𝘭. 𝘛𝘰 𝘢𝘷𝘰𝘪𝘥 𝘰𝘶𝘵𝘢𝘨𝘦𝘴, 𝘴𝘺𝘴𝘵𝘦𝘮𝘴 𝘯𝘦𝘦𝘥 𝘵𝘰 𝘦𝘷𝘰𝘭𝘷𝘦 𝘵𝘩𝘳𝘰𝘶𝘨𝘩 𝘱𝘢𝘳𝘢𝘭𝘭𝘦𝘭 𝘷𝘦𝘳𝘴𝘪𝘰𝘯𝘴, 𝘤𝘰𝘯𝘵𝘳𝘰𝘭𝘭𝘦𝘥 𝘳𝘰𝘭𝘭𝘰𝘶𝘵 𝘢𝘯𝘥 𝘳𝘦𝘷𝘦𝘳𝘴𝘪𝘣𝘭𝘦 𝘴𝘵𝘦𝘱𝘴. 𝘐𝘯 𝘰𝘯𝘦 𝘳𝘦𝘵𝘢𝘪𝘭 𝘱𝘳𝘰𝘫𝘦𝘤𝘵, 𝘸𝘦 𝘩𝘢𝘥 𝘵𝘰 𝘮𝘰𝘷𝘦 𝘵𝘰 𝘢 𝘯𝘦𝘸 𝘥𝘢𝘵𝘢 𝘮𝘰𝘥𝘦𝘭 𝘧𝘰𝘳 𝘳𝘦𝘤𝘰𝘮𝘮𝘦𝘯𝘥𝘢𝘵𝘪𝘰𝘯𝘴. 𝘐𝘯𝘴𝘵𝘦𝘢𝘥 𝘰𝘧 𝘢 𝘧𝘶𝘭𝘭 𝘴𝘸𝘪𝘵𝘤𝘩, 𝘸𝘦 𝘪𝘯𝘵𝘳𝘰𝘥𝘶𝘤𝘦𝘥 𝘪𝘵 𝘢𝘭𝘰𝘯𝘨𝘴𝘪𝘥𝘦 𝘵𝘩𝘦 𝘦𝘹𝘪𝘴𝘵𝘪𝘯𝘨 𝘴𝘺𝘴𝘵𝘦𝘮, 𝘮𝘪𝘨𝘳𝘢𝘵𝘦𝘥 𝘥𝘢𝘵𝘢 𝘪𝘯 𝘵𝘩𝘦 𝘣𝘢𝘤𝘬𝘨𝘳𝘰𝘶𝘯𝘥 𝘢𝘯𝘥 𝘳𝘦𝘥𝘪𝘳𝘦𝘤𝘵𝘦𝘥 𝘵𝘳𝘢𝘧𝘧𝘪𝘤 𝘴𝘵𝘦𝘱 𝘣𝘺 𝘴𝘵𝘦𝘱.’ 𝗧𝗵𝗲 𝘁𝗮𝗸𝗲𝗮𝘄𝗮𝘆: outages are the result of all-at-once changes. Teams can avoid them by designing changes so old and new can coexist, removing the single point of failure before any switch happens.

𝗧𝗵𝗲𝗿𝗲 𝗶𝘀 𝗻𝗼 𝘂𝗻𝗶𝘃𝗲𝗿𝘀𝗮𝗹𝗹𝘆 '𝗿𝗶𝗴𝗵𝘁' 𝗱𝗲𝗹𝗶𝘃𝗲𝗿𝘆 𝗺𝗼𝗱𝗲𝗹.  There is only the one that fits how your priorities. A simple way to choose: 𝗙𝗶𝘅𝗲𝗱 𝗣𝗿𝗶𝗰𝗲 works best when scope is clearly defined, timelines are predictable and dependencies are known in advance. 𝗧𝗶𝗺𝗲 & 𝗠𝗮𝘁𝗲𝗿𝗶𝗮𝗹 fits when scope evolves, priorities shift and decisions are made continuously during delivery. 𝗧𝗲𝗮𝗺 𝗘𝘅𝘁𝗲𝗻𝘀𝗶𝗼𝗻 is effective when you already have internal ownership, established workflows and need additional capacity inside your team. Our principles remain consistent across all models: - no upfront fees - pay only for work delivered - the ability to pause or adjust without heavy renegotiation - a clear legal and compliance baseline from day one In most cases, the right model aligns naturally with how your team is already set up to work.

𝗧𝗵𝗶𝘀 𝗶𝘀 𝗮 𝗴𝗿𝗲𝗮𝘁 𝗲𝘅𝗮𝗺𝗽𝗹𝗲 𝗼𝗳 𝘄𝗵𝗮𝘁 𝗵𝗮𝗽𝗽𝗲𝗻𝘀 𝘄𝗵𝗲𝗻 𝗮 𝗽𝗹𝗮𝘁𝗳𝗼𝗿𝗺 𝗼𝘂𝘁𝗴𝗿𝗼𝘄𝘀 𝗶𝘁𝘀 𝗼𝗿𝗶𝗴𝗶𝗻𝗮𝗹 𝘀𝘁𝗿𝘂𝗰𝘁𝘂𝗿𝗲. As the client’s platform expanded across regions and product areas, growth started to outpace the system’s ability to handle change. New analytics and AI requirements were layered on top of legacy components, creating delivery pressure and architectural bottlenecks. To address this and bring structure to parallel streams, we focused on several core areas: 𝗢𝗽𝗲𝗿𝗮𝘁𝗶𝗼𝗻𝗮𝗹 𝘄𝗼𝗿𝗸𝗳𝗹𝗼𝘄𝘀 𝗮𝗻𝗱 𝗮𝗻𝗮𝗹𝘆𝘁𝗶𝗰𝘀 We streamlined work with large academic catalogues through bulk operations, improved search and reporting. 𝗔𝗜 𝗿𝗲𝗮𝗱𝗶𝗻𝗲𝘀𝘀 𝗮𝗻𝗱 𝗱𝗮𝘁𝗮 𝗽𝗿𝗼𝗰𝗲𝘀𝘀𝗶𝗻𝗴 Content and student data were restructured for AI use, supported by scalable transformation and migration pipelines. 𝗗𝗮𝘁𝗮 𝘀𝘆𝗻𝗰𝗵𝗿𝗼𝗻𝗶𝘀𝗮𝘁𝗶𝗼𝗻 We aligned data flows across services to enable continuous changes in a live environment. 𝗤𝗔 𝗮𝗻𝗱 𝗽𝗿𝗼𝗱𝘂𝗰𝘁 𝗱𝗲𝘀𝗶𝗴𝗻 Ongoing QA and product design ensured that new functionality could be introduced without degrading user experience. As a result, the client was able to move forward with new initiatives across analytics, AI and core platform changes without being constrained by the system.  The collaboration remains ongoing, with continued contribution to the platform’s development and long-term scalability. 𝗪𝗵𝗮𝘁 𝗶𝘀 𝗵𝗼𝗹𝗱𝗶𝗻𝗴 𝘆𝗼𝘂 𝗯𝗮𝗰𝗸 𝗳𝗿𝗼𝗺 𝗺𝗼𝗱𝗲𝗿𝗻𝗶𝘀𝗮𝘁𝗶𝗼𝗻 𝘁𝗼𝗱𝗮𝘆?

𝗔 𝗯𝗶𝗴 𝗼𝗹𝗱 𝗯𝗮𝗰𝗸𝗲𝗻𝗱 𝘀𝘆𝘀𝘁𝗲𝗺. It’s a system no one wants to touch, but over time everything starts to depend on it. Changes get delayed, dependencies grow and even small updates require coordination across teams. In one of our recent projects, the product kept growing, but priorities started to compete.  So we had to operate across three parallel tracks: modernizing a legacy backend, redesigning the UI across web and mobile and supporting ongoing delivery. By focusing on where the system was constrained, we: - ramped up within weeks - reduced dependency on bottleneck zones - moved changes forward without losing control over the codebase and data Left alone, legacy systems don’t become safer. They become harder and more expensive to change. Worth fixing before it slows you down.

A day for family and a quiet pause. Happy Easter from Altabel Group!

This recommendation comes from our work with 𝗦𝗮𝗺𝗮𝗻𝗧𝗿𝗲𝗲 𝗠𝗲𝗱𝗶𝗰𝗮𝗹, a Belgium-based medtech company developing surgical imaging solutions. It is the result of years of continuous work, not a single delivery. Since 2017, the collaboration has grown into 6 projects across imaging systems, DICOM communication, desktop applications and backend platforms. Several of these systems are still evolving today, while the team grew to 15+ specialists working as part of the client’s team. During this time, the systems we worked on became part of a product that moved from early development to regulatory approval. The product also demonstrated measurable clinical outcomes, including a 67% reduction in reoperation rates and an 80% reduction in tissue assessment time. 𝗙𝘂𝗹𝗹 𝗰𝗹𝗶𝗲𝗻𝘁 𝗿𝗲𝘃𝗶𝗲𝘄 𝗮𝗻𝗱 𝗽𝗿𝗼𝗷𝗲𝗰𝘁 𝗱𝗲𝘁𝗮𝗶𝗹𝘀: https://lnkd.in/d8mVC4fe