Paul Preiss’ Post

Egnyte puts AEC AI Agents to work - extracting details from specification files and delivering guidance for building code compliance https://lnkd.in/eSm-Kmz3 #AI #Architecture #AEC #buildingcodes #specifications

There has been considerable discussion around "practical AI" as the appetite for incorporating AI into workflows continues to rise. Last week's announcement from Egnyte is a prime example of what practical AI looks like. Our new AEC AI Agents are specifically designed to solve challenges facing architecture, engineering, and construction firms on a daily basis by delivering AI guidance on lengthy, complex files. Thanks to Greg Corke for the great writeup!

New - Our paper "On Architectural Tactics for Resource-Constrained and Safety-Critical AI-Based Systems" was just presented at the 51st Euromicro Conference Series on Software Engineering and Advanced Applications (#SEAA) 2025 (Salerno, Italy). You can find the paper here: https://lnkd.in/e4KjG_d3 >> The goal was to identify the state-of-the-art applied architectural tactics and quality attributes to AI-based systems within our area >> Tactics are low-level building blocks of a system architecture, each influencing quality attribute fulfillment - making the tactic selection and balancing of attributes crucial steps in the architecture design process >> We conducted a Systematic Mapping Study based on 217 peer-reviewed publications Our work highlights 150 identified architectural tactics across 19 categories, providing a detailed overview that contributes to the development of safe and resource-friendly AI-based systems. Special Thanks to my supervisor, Marco Kuhrmann, for his valuable input, support, and guidance! No access to Springer? Just contact us to get the author's version. #softwarearchitecture #ai #tactics #tinyml #softwaredevelopment Philipp Straub, Christian Decker, Marco Kuhrmann at the Herman Hollerith Zentrum (HHZ).

Discover how AI is transforming architecture with intelligent designs, faster construction, and groundbreaking creativity. https://lnkd.in/dB3c_YnQ #AIRevolution #TechInArchitecture #AITrends2025 #DesignWithAI #SmartCities #ArchitectureLovers #FutureDesign #BuildingTheFuture #ConstructionTech #InnovationInDesign

We are using text-generative models like GPT, Llama and Gemini in almost all our day-2-day activities. But did we stop thinking what happens under the hood and how it works? These models are based on Transformer architecture, which was first introduced at 2017. The brake down of this architecture can be a bit hard to explain, but there is a great visualization (credit to: https://lnkd.in/dnyivbv5) that helps understanding the different phases and calculation along the way and thought to share with you: https://lnkd.in/dcggCpz9 I hope that can help make some order :) If not, feel free to reach out to discuss.

What will an architecture studio look like in 2035? AI will accelerate design, but meaning still comes from people culture, context, lived experience. As architects, how do we guide intelligent tools without losing our voice? #Architecture #ArtificialIntelligence #FutureOfWork #DesignInnovation #HumanCreativity #Architectureandai

Rajeev Bhargava’s latest article contrasts two paradigms of agency. The traditional Outcome → Calculation → Execution model has powered human civilization with predictability and control. But in today’s complex environments, a new model: Intent → Causality → Control is emerging, where systems adapt in real time and intelligence is embedded within. This shift raises deep questions about responsibility, governance, and trust in autonomous systems. It also points toward hybrid architectures that blend the stability of established frameworks with the adaptability of emergent reasoning. Read Full Article: https://lnkd.in/gsfy_vzn

🚀 From LLMs to Intelligent Agents! In this deck, I walk through: 🔹 Introduction & overview 🔹 LLM architecture & capabilities 🔹 Evolution to agents & core components 🔹 Real-world applications & use cases 🔹 Future outlook & challenges Excited to share insights on how AI is shaping the future of work and innovation. 🌐✨ 👉 What part of this journey interests you the most — architecture, applications, or the future outlook?

🚀 From LLMs to Intelligent Agents! In this deck, I walk through: 🔹 Introduction & overview 🔹 LLM architecture & capabilities 🔹 Evolution to agents & core components 🔹 Real-world applications & use cases 🔹 Future outlook & challenges Excited to share insights on how AI is shaping the future of work and innovation. 🌐✨ 👉 What part of this journey interests you the most — architecture, applications, or the future outlook?

Lightbulb moment: LLMs are pure functions 💡 Unless you are involved in building a foundational model (FM), the single most important thing that matters when building AI applications is how you 'engineer' everything around your chosen FMs and how you evaluate and improve the performance of your 'engineered' architecture (evals).