Sebastian Gauert

DATA IS WHERE IT'S AT! Well, that's been true for donkeys' years. Mr. Hollerith digitized data in 1889 then invented punch cards in 1911, thereby putting the "M" in IBM. Since then there have been too many "digital transformation" revolutions to count. Be that as it may . . . this is going to be ORCE (One Really Cool Event) - and I recommend anyone with an interest in modern approaches to data to have a look! BFH Technik und Informatik

6

Code assistant: software developers are working with AI-supported systems According to a Gartner study, around 75 percent of software developers in companies will be using AI-supported code assistants by 2028. At the beginning of 2023, the forecast was still below 10 percent. According to the survey, 63 percent of companies are already testing, using or have already used AI-supported code assistants. The advantage is that AI-supported code assistants allow more functions that go beyond code generation and completion. They are collaborative assistants that improve the efficiency of developers by stimulating brainstorming and improving code quality. This enables developers to continuously develop their skills in various frameworks. The opportunities provided by AI-powered code assistants lead to higher job satisfaction and bonding, reducing the costs associated with staff turnover. Software development leaders need to determine the ROI and build a business case when scaling their AI-powered code assistant rollouts. Traditional ROI frameworks direct executives to metrics that focus on cost reduction. However, this limited perspective does not capture the full value of AI-powered code assistants. #AI #GenAI #Developer #Cloud #CloudFirst https://lnkd.in/eNJtS5NT

38

3 Kommentare

❓❓ What can I believe ❓❓ The rise of (generative) #AI has opened tons of opportunities for us to ✅ augment our intelligence, ✅ get decision support, ✅ get help in discovering new insights and ✅ better interact with employees and consumers. However, unfortunately is has also opened a lot of opportunities for ❌ mis-use of the technology (which has happened to almost every technology before). #Desinformation is only one very severe possibility where AI can be used by people who deliberately want to deceive us 🤥, manipulate us 😮 and even want to influence elections 😤. Therefore, we are more than happy to join the "Bavarian Alliance against Desinformation" alongside many other tech and industry companies. Thanks to the Bayerisches Staatsministerium für Digitales and the Bayerisches Innenministerium for starting this initiative. Especially this year with so many elections - but in general always - it's really important to fight false information. More information here 👇 https://lnkd.in/eDkh_ycp #generativeAI #leadership #democracy Rolf Löwisch IBM

30

1 Kommentar

One of the biggest challenges in planning and reporting is the targeted preparation of statements to the management level based on the data and, above all, the comments. Controllers spend hours and days sifting through data and text, processing and summarizing reports to enable their superiors to make intelligent decisions. The problem: while numerical values are added and subtracted by their planning-software and the totals can be understood by their components, the problem with comments is the interpretation, language, quality and how to consolidate them and work them to a presentable message. You don't want to simply consolidate the comments or pick out individual ones and neglect others. In the end, business decisions should be made on the basis of all applicable data and comments and not just a fraction of it. To find a solution to this issue, what could be more obvious than to make use of modern generative AI methods? In our specific case with IBM Planning Analytics, we use our Jupyter Lab (GMC² PACO HUB) and connect the TM1 data model with freely available LLMs such as Gemma, Llama3, Mixtral, etc. via TM1py - but of course watsonx or the current GPT-4o are no problem either. There we generate management summaries automatically, or at the push of a button from PAW, across any level of the multidimensional data cube. The highlight: it is completely irrelevant in which language the comments were written and in which language I want the comments to be displayed - the translation can be taken over at the same time and populated on another layer of my TM1-model. And all of this neatly integrated with IBM Planning Analytics and the use of open source software. This makes it easy to get a taste of the world of generative AI with a focus on planning and reporting. We discuss this in our latest blog-entry in more detail. https://lnkd.in/e_sx2V3k The text of the blog is in german language, but I guess you already know of some neat-translation-solutions. GMC² GmbH IBM #ibmplanninganalytics #planninganalytics #tm1 #genai #managementsummary #predictiveanalytics #machinelearning

18

1 Kommentar

Lies hier den neusten Blog-Beitrag der arelium😊! Data Mesh in Microsoft Fabric

1