January 2026

Why Shanghai’s medical AI showdown reveals more about system architecture than clinical superiority The headline writes itself: AI defeats human doctors in medical diagnosis, delivering results in under two seconds versus 13 minutes. But beneath this dramatic 390x speed differential lies a far more nuanced story about specialized model architecture, the evolving role of clinical decision support systems, and the critical distinction between diagnostic speed and comprehensive patient care. The Technical Architecture Behind the Victory Shanghai AI Lab’s Gastrointestinal Multimodal […]

Ver mais

Photo by Barrett Ward on Unsplash This project was presented as the final project to obtain the Founderz AI Business School + Microsoft certificate in AI and Innovation 2025. Summary In contemporary logistics, punctuality has emerged as a fundamental pillar for customer loyalty and business competitiveness. However, a significant part of supply chains continue to operate under reactive paradigms, limiting themselves to managing delays only when they have already materialized. This project addresses this inefficiency by developing a comprehensive […]

Ver mais

I lost one Saturday trying to make existing SVG plugins work with Rslib. Most were Vite-only, some were outdated, and none fit Svelte or Vue properly. So I wrote two plugins instead: @avatune/rsbuild-plugin-svg-to-svelte and @avatune/rsbuild-plugin-svg-to-vue This post is basically why they exist. The Setup I’m building Avatune, an avatar component library that supports React, Vue, Svelte, and Vanilla JS. Same SVG assets, different framework outputs. The architecture looks like this: raw SVG files for avatar parts (heads, eyes, […]

Ver mais

Your Employees Are Feeding Secrets to AI — And You Don’t Even Know It The $670,000 Mistake You’re Probably Making Right Now While you’re reading this, someone in your organization is pasting confidential data into an AI tool. They’re not malicious. They’re not stupid. They’re just trying to finish a report 10 minutes faster. And that’s exactly why Shadow AI is your biggest security nightmare in 2025. Shadow AI — the unauthorized use of AI tools by employees […]

Ver mais

Large language models (LLMs) exhibit impressive in-context learning (ICL) capabilities, yet the quality of their predictions is fundamentally limited by the few costly labeled demonstrations that can fit into a prompt. Meanwhile, there exist vast and continuously growing amounts of unlabeled data that may be closely related to the ICL task. How to utilize such unlabeled data to provably enhance the performance of ICL thus becomes an emerging fundamental question. In this work, we propose a novel augmented […]

Ver mais

Agentic AI systems act autonomously across enterprise environments, making traditional perimeter-based security ineffective. Zero Trust treats AI agents as independent actors with identities, enforcing continuous verification, least-privilege access, and contextual controls across APIs, systems, and data. While powerful, Zero Trust must be implemented thoughtfully to avoid policy sprawl, blind spots, and legacy gaps. When done right, it enables organizations to scale autonomous AI safely—without sacrificing speed, innovation, or trust.

Ver mais

The failure wasn’t technical When you’re responsible for rolling out AI inside a company, it’s easy to focus on the visible parts. Models. Connectors. Retrieval. Permissions. Latency. That’s the work everyone sees. But that’s not why my rollout struggled. It didn’t fail because the system was broken. It failed because I was designing AI for teams instead of with them. At Nextdoor, our early Work AI rollout stalled even though the technology was solid. Accuracy was fine. Performance […]

Ver mais

Hello Everyone, Since the last few months, I have been studying about world models and along side built a library for learning, training and building new world model algorithms, pytorch-world. Added a bunch of world model algorithms, components and environments. Still working on adding more. If you find it interesting, I would love to know your thoughts on how I can improve this further or open for collaboration and contributions to make this a better project and useful […]

Ver mais

Personalized Federated Learning (PFL) enables collaboratively model training on decentralized, heterogeneous data while tailoring them to each client’s unique distribution. However, existing PFL methods produce static models with a fixed tradeoff between accuracy and efficiency, limiting their applicability in environments where inference requirements vary with contexts and resource availability. Early-exit networks (EENs) offer adaptive inference by attaching intermediate classifiers. Yet integrating them into PFL is challenging due to client-wise heterogeneity and depth-wise interference arising from conflicting exit objectives. […]

Ver mais