Sudeep Das shares how DoorDash shifts from legacy one-shot predictions to an agentic recommendation platform. He discusses leveraging language-native consumer memory, RQ-VAE semantic IDs for...

FedWorld提出一种作用域感知的联邦世界模型协议，解决多Agent知识共享冲突。它通过交换结构化抽象规则而非参数或轨迹，基于证据强度决定规则共享、集群专属、私有或未决，避免负迁移。实验显示在τ-bench和ALFWorld上减少状态回退、提高成功率，但存在规模有限、对齐准确性等局限。

Microsoft released a dedicated AI Gateway tier of Azure API Management in public preview, with a control plane built around models, MCP servers and tools rather than APIs. It fronts Foundry,...

Google shipped AlloyDB AI functions GA with a proxy model architecture that trains a lightweight local model from LLM outputs, then runs queries at database speed without external calls. Smart...

Naomi Saphra discusses 5 rules governing language model behavior, breaking down why LLMs act like populations rather than individuals. She explains how tokenization creates strange semantic blind...

Monzo recently redesigned its data warehouse to support more than 100 teams working on over 12000 dbt models. Introducing a so-called "meshy" approach, Monzo cut warehouse costs by about 40% and...

NVIDIA has announced a new family of open models called NVIDIA Ising, designed to address quantum processor calibration and quantum error correction. These are two of the main engineering...

GitHub will use Copilot interaction data from Free, Pro, and Pro+ users to train AI models starting April 24, opting in by default. Collected data includes code snippets, inputs, outputs, and...

In this episode, Shweta Vohra and Joseph Stein explore what changes when software systems start planning, acting, and making decisions on their own. The conversation distinguishes truly agentic...

In the lab, your AI model might seem perfect, but the real world is often where it breaks.

Six business models piloted in Asia offer lessons for global leaders in any market—and just might be the key to the next wave of sustainable growth.

For most small- and medium-sized business leaders, the question about AI has shifted. While it used to be “Should we use AI?”, it’s now “Where should we run it?”

As UX and product leads, we sit right at the intersection of user needs, business goals, and technical constraints. This is exactly where most AI products fail. I’ve seen this pattern repeatedly...