ING Australia

Something worth knowing if you own a home in NSW. From mid-2026 sellers will be able to voluntarily include their home's NatHERS energy rating in property listings. I know "voluntary disclosure" doesn't sound like a big deal. But the ACT has had mandatory energy disclosure for years. Europe has had it for over a decade. The research consistently shows the same thing, energy efficient homes sell faster and command a premium once buyers have information to act on. The homeowners who get their homes rated and act on the results before this becomes mainstream are going to be in a better position. That's why I built the homeowner side of EnBright. Not just to lower bills, though it does that too. To help homeowners make a decision that pays off now and at the point of sale. enbright.com.au #HomeUpgrades #EnergyBills #GreenHomes #Electrification #SustainableLiving #NatHERS

9 months ago, I made a post wondering what a major shift around CyberArk could mean for the PAM world and whether the industry was heading toward consolidation. Now, with Palos move to rebrand CyberArk “Idira” identity, it honestly feels like we’re watching the IAM and PAM industry evolve in real time. And I keep asking myself: Was PAM ever going to remain just “PAM”? For years, privileged access management sat in its own lane. Separate teams. Separate tooling. Separate conversations. But today the boundaries are getting blurry very quickly. PAM, IGA, MFA, cloud permissions, secrets management, machine identities, Zero Trust, AI agents… everything is slowly converging around one thing: Identity. Not identity as a login screen. Identity as the actual control plane of security. Humans, apps, workloads, APIs, automation, AI agents, all need governance, trust, visibility, and controlled privilege. That is why this shift feels bigger than a rebrand to me. It feels like the industry is preparing for the next era where identity security becomes the centre of enterprise security architecture, not just another security domain. Maybe this is strategic positioning. Maybe this is preparation for AI-driven environments. Or maybe this is simply where the industry was always going to end up. Either way, it’s a very interesting time to be working in IAM and PAM. Curious to hear from other peers; Are we seeing the next evolution of PAM happening right now? #CyberSecurity #IAM #PAM #IdentitySecurity #CyberArk #IGA #AI #IdentityManagement

After 9 years and 8 months… wow, almost a decade. If ING were a TV series, this would be my final season—and it’s time for my character to take a bow. 🎬🧡 ING has been more than just a workplace. It’s been my home, my safe space. It has taught me so much—the good, the tough, and everything in between. I’ve had the privilege of working with incredible people: inspiring leaders (and yes… a few bosses 😄). Like any home, it wasn’t perfect—I’ve loved it, and at times, I’ve hated it—but every moment shaped me. This journey opened doors I never imagined and gave me experiences I’ll carry forever. What I’ll treasure most are the people—colleagues and acquaintances who became friends, and friends who became family—something I never expected to find at work. So with that, I say thank you. Truly. The orange lion will always be part of my story—my adventure. 🦁 To everyone I’ve shared this journey with… this isn’t goodbye, just a heartfelt “see you around.” ✨ Toffer signing off 🫡

I’m pleased to share our latest manuscript, “Quantum-Inspired Geometric Classification with Correlation Group Structures and VQC Decision Modeling.” This project represents an important milestone for me—not only from a research perspective, but also as a testament to the teamwork, perseverance, and intellectual curiosity that shaped its development. I would like to express my sincere appreciation to my coauthors Nishikanta Mohanty (Ph.D.) , Bikash Kumar Behera , and Badshah Mukherjee for their dedication, insightful collaboration, and steady support throughout this work. It has truly been a rewarding experience to build this together. 🙏 What makes this research particularly interesting is its geometry-centered approach to classification, moving beyond traditional methods that depend primarily on probabilistic decision boundaries. The framework integrates several complementary components: Correlation Group Structures (CGR) that structure features into correlation-sensitive neighborhoods. Efficient SWAP-test-based overlap estimation to capture both Euclidean-style and angular similarity information. A deterministic Fusion Score mechanism that provides an interpretable and computationally light classification strategy for small-to-medium datasets. D-distance contrastive representations combined with a selective Variational Quantum Classifier (VQC) stage designed to address large-scale and highly imbalanced problems such as fraud detection. From my perspective, the key contribution lies in how these elements are unified into a hybrid quantum-classical learning pipeline that remains interpretable, adapts to varying data conditions, and emphasizes geometric similarity as the foundation of decision-making rather than relying solely on posterior probability estimation. In practice, the approach compares samples to class medoids, extracts geometric evidence derived from overlap measures, and activates the VQC component only when additional nonlinear refinement is beneficial. A notable point is that this framework differs from conventional quantum kernel-based classifiers. While many kernel approaches use overlap values to construct kernel matrices for subsequent classical learning, our method leverages overlap directly to form transparent geometric decision signals. This design makes the system more prototype-oriented, interpretable, and modular. The study demonstrates competitive results on benchmark datasets including Heart Disease, Breast Cancer, and Wine Quality, and further extends to Credit Card Fraud detection using a D + VQC pipeline tailored for extreme class imbalance and performance evaluation at specific operating points. I’m thankful for the opportunity to contribute to the evolving field of quantum-inspired machine learning, particularly in ways that aim to balance theoretical rigor with real-world applicability. 🚀 Preprint available here: https://lnkd.in/eZd2hUS3 #MachineLearning #QuantumComputing #ArtificialIntelligence