Wei Wang

Introducing the M5 The M5 chip brings significant advancements in performance and efficiency across various applications. 1. Massive GPU / AI performance boost The M5 features a next-generation 10-core GPU architecture, with each core equipped with an integrated Neural Accelerator. This results in over 4× the peak GPU compute performance for AI tasks compared to the M4. 2. Stronger CPU and enhanced Neural Engine With up to a 10-core CPU (6 efficiency cores + 4 performance cores), the M5 offers approximately 15% faster multithreaded performance over the M4. Additionally, the Neural Engine has been upgraded to 16 cores, enhancing performance for on-device AI and intelligent features. 3. Higher memory bandwidth / unified memory improvements The M5 provides around 153 GB/s of unified memory bandwidth, which is approximately 30% more than the M4. This unified memory architecture allows all components (CPU, GPU, Neural Engine) to access a larger memory pool, enabling larger AI models to run entirely on-device. 4. Graphics & rendering enhancements The improvements in the GPU also boost graphics performance, including ray tracing capabilities. Apple claims up to a ~45% uplift compared to the M4 in ray-traced workloads, enhancing visual fidelity, rendering speed, and smoothness in demanding 3D and graphics applications. 5. Deployment in new hardware + energy efficiency / environmental goals The M5 will power new devices, including the 14-inch MacBook Pro, iPad Pro, and upgraded Vision Pro, all of which are available for pre-order.

7

From rockets to AI. Nine years after the original NVIDIA DGX-1 handoff, Jensen Huang delivered a brand-new DGX Spark to Elon Musk at Starbase. The DGX Spark is a petaflop powerhouse built for creators, researchers, and developers — a desktop-sized supercomputer with five times the performance of the first DGX-1. 🚀✨ Filmed at SpaceX rocket factory in Texas, this moment captures a conversation between the two visionaries. ➡️ https://bit.ly/4o7oqGw 📹 https://bit.ly/42GxidS

26

Yesterday at NVIDIA Live at #CES2026 CEO Jensen Huang shared how AI is transforming every domain and device, redefining the future of computing. He unveiled the Rubin platform, a new extreme-scale AI supercomputer architecture now in production, alongside major advances in open models and autonomous driving. 🗞️ Catch up on all the announcements: https://bit.ly/45l899Z

10

🤖 Improving #supplychain operations through intelligent #robotics. Otto Group is taking intelligent #automation to the next level — with #NVIDIA. #OttoGroup announced a collaboration with NVIDIA to deploy an AI-powered Robotic Coordination Layer across 120 logistics locations, enabling centralized coordination of robotic systems across the retail supply chain. This initiative positions Otto Group as a first European mover in #retail to implement network-wide robotic orchestration at scale. Built with NVIDIA #AI and delivered on Google Cloud, the first blueprint is live at Hermes Fulfilment’s logistics center in Löhne, with Reply supporting the initial implementation. #Gogle #Cloud #GCP

11

Yesterday at NVIDIA Live at CES 2026 CEO Jensen Huang shared how #AI is transforming every domain and device, redefining the future of computing. He unveiled the Rubin platform, a new extreme-scale AI supercomputer architecture now in production, alongside major advances in open models and autonomous driving. 🗞️ Catch up on all the announcements: https://bit.ly/3Yv6b2L

8

We’re excited to announce that the code for our paper, “EpiCache: Episodic KV Cache Management for Long Conversational Question Answering,” is now open-sourced at https://lnkd.in/gtfqhjMU EpiCache introduces an episodic approach to managing key–value (KV) caches in large language models, enabling longer, more coherent, and more efficient conversations without the prohibitive memory and compute costs of traditional caching. Our approach dynamically decides which conversational “episodes” to retain or discard — striking a balance between context preservation and resource efficiency. This work contributes to advancing scalable long-context reasoning and efficient LLM inference, especially for interactive and multi-turn conversational tasks. Read the paper for technical details, and explore the code to experiment with EpiCache in your own conversational systems. Minsoo Kim, Han-Byul Kim,Richa Dixit,Minsik Cho https://lnkd.in/gtfqhjMU

155

New from O’Reilly + Redis: Managing Memory for AI Agents 📘 AI agents are only as good as their memory. This new report explores how short-term, long-term, and persistent memory shape agent performance—and how Redis provides real-time, scalable architectures for production-ready AI. 📖 Read the full report: https://lnkd.in/g7t--57p

50

1 Comment

When Steve Huynh was a principal engineer at Amazon, meetings began with a "study hall." Amazon had a "reading culture" even among engineers, Huynh recently told the Pragmatic Engineer podcast, speaking of his time at the tech giant. Employees frequently drafted six-page memos, which they shared with the company to update progress and demonstrate new projects. Huynh, who said the company's embrace of writing and reading the memos was part of its "secret sauce," said employees' writing was often constrained to the format, whether it was a business strategy or press release. Huynh started at Amazon in 2006, only a few years after the company turned its first profit and while Jeff Bezos was at the helm. Read more about how Bezos instilled this culture of memo-writing from the top down on Business Insider: https://lnkd.in/gBqbEf_j (Credit: Daniel Berman) #amazon #reading #jeffbezos

216

2 Comments

After 8 rewarding years, I bid goodbye to Amazon. I strongly believe that successful companies are built with strong culture at their core, and Amazon’s Leadership Principles are some of the most recognizable ones. I'm grateful that Amazon is where I bootstrapped my career. As I reflect on my journey, I also want to share some personal takeaways that shaped my growth. And of course, in the spirit of “if everything is important, then nothing is”, here goes with my top three. 1. 𝐖𝐨𝐫𝐤 𝐨𝐧 𝐰𝐡𝐚𝐭 𝐦𝐚𝐭𝐭𝐞𝐫𝐬 - You should work on: 1) Thing that matters to your team/org/company. If you are not clear, clarify with your leaders and help define one if needed. What matters? If you work backward from the needs of your customers (those who use your product/service), you will get to it. 2) Thing that matters to you. Ask yourself what motivates you and what can help you learn. Find opportunities that push you closer to what you want to achieve in the long run. If you’re lucky, you get to work on 1 & 2 at the same time. Be conscious of what you work on. 2. 𝐋𝐞𝐚𝐫𝐧 𝐛𝐲 𝐨𝐛𝐬𝐞𝐫𝐯𝐢𝐧𝐠 - Over the years, I’ve been fortunate enough to work with a good set of top talents. I learned the most and the fastest by simply observing how they made decisions at high-stakes moments, how they brought order into chaos, how they knew when to move fast and when to slow down, and how they behaved under high pressure, etc. Try to reason about and internalize those behaviors, and adapt to similar situations you face. Like the phrase “Follow the smart money” in the investment world, go identify and learn from the greats. 3. 𝐂𝐨𝐧𝐟𝐢𝐝𝐞𝐧𝐜𝐞 𝐠𝐨𝐞𝐬 𝐚 𝐥𝐨𝐧𝐠 𝐰𝐚𝐲 - Career is made up of ups and downs. Confidence is about not shying away from advocating for your work and celebrating your wins. Confidence is also about resilience - learning from setbacks, carrying the lessons forward, and moving on. Confidence is about showing up with energy, curiosity, and enthusiasm in a way that inspires those around you to produce positive results. Be confident, and just have fun! P.S. None of the above, in my opinion, will be replaced by AI anytime soon. :) #growth #day1

869

25 Comments

I owe a lot to Amazon for the first 5 years of my career. That time gave me a base I still lean on every week: ➤System design habits – Thinking about traffic patterns, failure modes, backpressure, blast radius. ➤Engineering discipline & Mentoring Others – Code reviews, testing, on-call, writing clear docs, closing the loop on incidents. ➤Ownership – If it shipped with your name on it, you stayed till it worked. All of that has helped me much more than any job title. After Amazon, I had to learn a different set of skills, like: - Shipping without a huge internal platform solves half the problems. - Dropping habits that slow you down when you are ten people & not ten thousand. - and many other small things you get used to over months. Both phases were useful in very different ways. So when I see takes like, “Nothing I learned at FAANG helped me build a startup.” I usually think of two questions: 1. How much did you really get to own there - A full project, or a summer intern ticket? 2. What did you choose to carry forward - Good reviews, design habits, writing skills, or only the brand on the badge. Big tech is not a magic stamp for greatness. Startups are not the only place where real work happens. Every environment leaves you with tools and with blind spots. The upside comes from what you keep, what you unlearn, and how you grow beyond the logo on your access card.

718

19 Comments

David Luan’s Amazon AGI SF Lab marks a significant evolution in Amazon’s AI strategy, distinguishing itself from the company’s traditional AI initiatives. While Amazon has historically focused on scalable infrastructure and consumer-facing products, the AGI SF Lab, under Luan’s leadership, is dedicated to developing autonomous AI agents capable of performing complex tasks across digital and physical environments. The lab’s flagship model, Amazon Nova Act, exemplifies this shift. It outperforms competitors like OpenAI and Anthropic on benchmarks assessing AI agents’ intelligence and aptitude, particularly in tasks involving web interactions . This model integrates reinforcement learning and draws inspiration from robotics to enhance decision-making capabilities, aiming to create AI agents that are dependable and capable of independent actions. Interested to learn more about the AGI SF Lab? Check out the team’s page here: https://lnkd.in/eWHQC7Ps

97

2 Comments

⚡ Exciting to see our Block-Max Pruning (BMP) technique in Infinity, an open-source AI-native database designed for LLM applications! In their latest VLDB paper, “Balancing the Blend: An Experimental Analysis of Trade-offs in Hybrid Search”, Hai Jin, Yingfeng Zhang, and co-authors present a rigorous evaluation of hybrid search architectures — combining full-text, sparse, dense, and tensor retrieval. To support efficient sparse vector search at scale, they’ve integrated BMP into Infinity’s SVS engine — a nice validation of our work on fast, top-k lexical retrieval. 🔗 BMP paper: https://lnkd.in/dsc33hGc 🔗 BMP code: https://lnkd.in/dxBxv225 🔗 Infinity: https://lnkd.in/ddRK5mbr 🔗 Hybrid Search paper: https://lnkd.in/dfBuDXmt Great to see ideas from traditional IR continuing to shape the next generation of retrieval infrastructure!

31

If you're a Senior Engineer looking to move up, the next role will likely be a Staff Engineer. And in this guide, Shruti shares tips from her own experience of getting promoted to Staff Engineer at PayPal and Slack. She talks about what Staff Engineers do (and how it's different from Seniors), why you might not be getting promoted, and how to take that next step. https://lnkd.in/g58dnFEG

103

3 Comments