Sign in to view Lee’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Lee’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Seattle, Washington, United States
Sign in to view Lee’s full profile
Lee can introduce you to 10+ people at Google
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
4K followers
500+ connections
Sign in to view Lee’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Lee
Lee can introduce you to 10+ people at Google
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
View mutual connections with Lee
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Sign in to view Lee’s full profile
or
New to LinkedIn? Join now
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
About
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
Articles by Lee
-
Paying it forward
Paying it forward
I’m coming up on 4 weeks at Google Cloud, and I wanted to share my views as I get my feet wet! I’m enjoying working…
22
Activity
4K followers
-
Lee Gates shared thisLee Gates shared thisGoogle Cloud named a leader for the 5th consecutive year! I am very stoked to see Google Cloud's big advances in this year's Gartner MQ for cloud infrastructure. Thank you Gartner for the thoughtful and rigorous work that goes into it. Deep gratitude to our engineering teams, field teams, partners, and customers for the trust you place in us. Here's Brad Calder's take on how a curated, workload-optimized approach is helping fuel our momentum. 👇🏼👇🏼 👇🏼Google is a Leader in Gartner Magic Quadrant for Cloud Infrastructure and Platform Services | Google Cloud BlogGoogle is a Leader in Gartner Magic Quadrant for Cloud Infrastructure and Platform Services | Google Cloud Blog
-
Lee Gates shared thisI had a few minutes reflecting on all the support I received over the past few years and realized I could help others considering career decisions...
-
Lee Gates posted thisI've joined Google Cloud to work on Product Management for performance and availability across the platform. Super excited to join and am happy to be working with a great team. Some technologies create incredible value. Past examples include telegraph, telephone, radio, tv, cable, internet, wireless communications, blockchain, etc. Public cloud is in that list. We are innovating and building new infrastructure and services to deliver incredible global impact. During these difficult times cloud has become an enabling technology platform for industry and social continuity. At the same time, cloud is becoming the foundational technology for many more disruptive innovations. I'm excited every day to work with great people and make this level of impact. As one of my favorite leaders often said, ONE TEAM!
-
Lee Gates shared thisNew on-demand OCI Performance Webinar! Intro to our performance program with methodology for benchmarking and performance analysis update. Michael Kane joins me to present OCI Dyn DNS, attacking the same problem from the edge. #oci #performance #benchmarks
-
Lee Gates shared thisWrapping up our 1st #OracleCloudInfrastructure Deep Dive Days (D3) in Tokyo! We had a great time with all of our partners in Japan and wrapped up with final sessions on Solution Architecture and Performance. #OracleCloudInfrastructure Great questions and excitement as we get ready to bring high performance and enterprise cloud to your Japan customers. #OraclePartners We'll be coming to a city near you this year with our D3 series. Can't wait? We've got a new page with training resources https://lnkd.in/gb3BaTp Frank Obermeier Kash Iftikhar James Stanbridge Shinji Taketsume Takahito Kawashima Arthur F. Tyde III Camillo Speroni Ed Omland Eli Schilling John Growdon Hiroshi Saito Ed Zou A. Arslan Munir Raja Mukherjee Brandon Thomas Julie Tung Tomohiro Hara Hisayuki Ishizumi Ani Fox Bochenkov Makoto Honda Toru Hayashi Masaki Hamura, Ph.D. Koji Minato Hiroaki Komeno Bo English-Wiczling Tatsuya Shiratori Gyota Kondo
-
Lee Gates liked this*cries in almquist shell calling ubus and uci*Lee Gates liked thisUnix nerds, you win. Forty years later, everything runs on shells, Python scripts, and cron jobs. The GUI was just a phase.
-
Lee Gates liked thisLee Gates liked thisLast week, I left Google Cloud for an exciting new opportunity (more on this later). It was a short stay, but a momentous time for Google’s AI and Cloud Infrastructure team, the foundation for Alphabet’s AI strategy, and home to 9 of the 10 top AI Labs. In 15 months, we launched two new TPU platforms, Ironwood and TPU 8, and a range of NVIDIA Blackwell-based machines including the B200, GB200, GB300, and RTX Pro 6000. We launched several groundbreaking Kubernetes training, inference, and agent capabilities, cementing Kubernetes as the AI orchestration platform. And we filled out a robust storage and networking portfolio with advancements like Google Cloud Storage Rapid, Managed Lustre, Hyperdisk ML, and the Virgo Network. On the CPU side, we launched our fourth generation compute instances across general purpose, memory optimized, storage optimized, network optimized, and HPC optimized, including Google’s own Axion Arm instances. I’m very proud of the team and the work we did to help customers and help the business grow from a $49 billion annual run rate to a staggering $80 billion! Thank you to Nirav Mark George and Drew for trusting me and the OPM team with the messaging and positioning. Thank you to the awesome product management team Souvik Jeff Salil Cheng Sabastian Mohan Alex Alex Shubha Nathan Akshay Sean Edward Allan Gurmeet Connor Yatharth Peter Pramod R. Benny Ari Venkat Rajeev Chanuka. And thank you to the great GTM, strategy, and marketing teams Flora Niamh Ali Alexander Eyvonne Jarrad Brian Neha Leonora Fazil Lillian Shiva. (I've just found the LinkIn mention limit!). Last but not least, thank you to my team: Jason, Sai, Chelsea, Bala, Moazzam, Alan, Lee, Sajal, Dave, Bobby, and Injae. We had a great run and I'm so proud of what we accomplished together. Google Cloud’s mission and full stack AI strategy is powerful and I’ll be excited to watch it continue to unfold.
-
Lee Gates liked thisLee Gates liked thisWelcoming Roy Kim to the Lightmatter Leadership Team! We are thrilled to announce that Roy Kim has joined Lightmatter as our new Vice President of Product. Roy is a seasoned leader in the AI infrastructure space, bringing over 15 years of experience from industry titans including Google, AMD, and NVIDIA. Most recently, as Director of AI Infrastructure Product Management at Google, Roy led the strategy for large-scale AI initiatives—giving him a unique, end-to-end perspective on how hyperscale AI infrastructure is designed, built, and deployed. At Lightmatter, Roy will lead our global product strategy and roadmap as we enter a new chapter focused on production at scale across hyperscale AI infrastructure. His expertise in aligning engineering, commercial, and customer needs will be instrumental as we continue to deliver industry leading products and redefine the future of data centers. Please join us in giving Roy a warm welcome to the team! We are excited to have his leadership as we scale our solutions to meet the explosive demand of the AI era. Read the full announcement here: https://bit.ly/4tDOMCS Nicholas Harris #LeadershipAnnouncement #AIInfrastructure #Photonics #Semiconductors #ProductManagement
-
Lee Gates liked thisDive DEEP into how Google Cloud and NVIDIA architect reliable rack scale NVL72 systems with this session 🔧 Then swing by the showcase floor to see A4X Max (NVIDIA GB300) in person. I’ll be there to answer any questions [..and snap pics for you with A4X 📸] Google Cloud Next '26 Google Cloud Partners NVIDIALee Gates liked thisHeading to Google Cloud Next this week? Join Pramod R., Deepak Narayanan (Nvidia) and me on Thursday 04/23 at 11:45am for a deep dive into building NVLink-connected, rack-scale systems. We’ll explore the system design and data center innovations required to enable A4X Max instances – powered by the Nvidia GB300 NVL72 – and consider the future A5X product line, powered by Nvidia’s Vera Rubin architectures. We will also discuss the Nemotron-3-Super model, which was pre-trained on the NVL72 system. We hope to see you there: BRK3-036, Thursday, 11:45 a.m., South-Seas B. https://lnkd.in/ghR6pKqU #GoogleCloud #GoogleCloudNext #GoogleCloudNext2026 #AIHypercomputer #ReliableInfra
-
Lee Gates liked thisLee Gates liked thisYou'll see lots of posts about Cloud NEXT sessions this week. Here's my reminder to come check out Edward Villalba, Michael Ciacco and I on Wednesday morning if you're free. We'll be on the 3rd floor in South Seas H. You can't accelerate agentic workloads if you don't also upgrade the automation in your cloud environments. A Live Demo, audience questions and a great customer use case. We hope to see you there.
-
Lee Gates liked thisLee Gates liked thisSoftware company product managers and architects won't want to miss this. I'm delighted to host an awesome panel of technical leaders including Gelu Ticala from Kinaxis, Benjamin Flast from MongoDB, Rohit Mendiratta from Palo Alto Networks, and Murray Stokely from Snowflake. We'll dig into how they've designed their software for maximum performance, cost-efficiency, and flexibility using some of Google Cloud's unique capabilities in #kubernetes, custom machine types, Hyperdisk storage pools, Axion Arm compute, and more. We'll also talk about how good architecture enables new products for new customer bases, and how they're designing for AI. Register now! https://lnkd.in/gsKptgQb Breakout 1 (BRK1-066): Architecting high-performance SaaS Thursday 8:00-8:45am South Seas F Check out other top infrastructure sessions: https://lnkd.in/gWW7HHdj #cloudcomputing #generativeAI #software #googlecloudnext #googlecloudnext26 Mark Nirav Drew Peder Neha Tanuj Salman Henri Amin Jason Saiganesh Will Richard Google BJ
-
Lee Gates liked thisI’m excited to share that my next chapter has started with Cox Automotive Inc. If you’ve known me for more than 30 seconds, you already know how deep my passion is for cars and the automotive industry. I couldn’t be more excited to bring my personal and professional passions together at Cox Automotive. Most companies only see a piece of the transaction they support, but Cox Automotive’s data foundation and portfolio allow them to optimize the entire journey, from the first time a customer starts their search to driving their new car off the lot. The automotive industry is at an inflection point. AI, shifting consumer expectations, and the digitization of historically complex transactions are converging in ways that will reshape how people buy and sell cars. Cox Automotive is leaning into this moment. By investing in people, building on strong platforms, moving faster with AI, and staying relentlessly focused on the customer, Cox Automotive is well positioned to define the next generation of automotive retail. I’m looking forward to working with an incredible team and contributing to what comes next. #AI, #LifeatCOX, #Automotive, #EcommerceLee Gates liked thisWe are excited to welcome Jeff Jones to Cox Automotive as SVP of Product and Engineering, Consumer and Marketing Solutions. Jeff joins at a pivotal moment as AI and rising consumer expectations continue to reshape how vehicles are bought and sold. He brings a deep understanding of eCommerce and digital shopping, along with a clear focus on end-to-end visibility across a complex transaction. That perspective helps us keep building tools that connect shopper interest to signed deals, supporting better experiences for consumers and the dealers that serve them. Read the full announcement here: https://bit.ly/4tkM3xW
-
Lee Gates liked thisThank you to everyone that took the time to talk with me during office hours. I really enjoyed and got a lot out of those conversations. I’m suspending office hours for a while until I have a better understanding of the time requirements of my new gig. Details next week. SOOOOO excited!!!!Lee Gates liked thisJeffrey's retirement but like..... (watch this space)
-
Lee Gates liked thisLee Gates liked thisToday I am not sharing the next technological trend with you. Today is about me. Starting April 1st, I'm officially a Vice President at Gartner! Joining Gartner was one of the best decisions I've made for my career. When I was 16 I had the opportunity to work on the first enterprise system of my career. It was my summer job and I had the opportunity to land my eyes on an HP Superdome (fun fact: it was the only generation already black but still based on PA-RISC instead of Intel Itanium). Almost 20 years later, after relocating to a different country when I was 22, here I am, Vice President Analyst of one of the organizations that employs brilliant people tasked to support our clients in achieving their goals through objective and actionable advice. And I was never one of these crazy guys, only focused on work: in the meantime I had the luck to meet my lovely wife and have with her our two beautiful daughters. What a ride! What I really like here at Gartner is the culture: we are encouraged to build on each other's knowledge and expertise. It is not a competitive sprint between different analysts, but is a group marathon that we run together, and where we support each other. The spirit of camaraderie between analysts is something I never saw before in my professional life. And…it works! It helps us to move our research forward and, most importantly, better serve our clients. For me it is awesome to enter a room with so many brilliant people and know already that I am not the smartest one there, so I get to learn constantly. I've learned a lot over the last three years, thanks to countless great chats with my fellow Gartner analysts, our clients and the vendors. And learning never stops here, I can’t wait to continue to learn, discuss, and move forward the IT industry. #gartner #CareerGrowth
Experience & Education
-
Google
***** ******* *******
-
*******
********* *********
-
******
**** ** *********** *********** * ****** ***** **************
-
********** ** *********** ********
*********** ********** ************ ******* *********** ********* ********** undefined
-
-
****** ***** **********
********** ****** ******** ******** ******* ** ***********
-
View Lee’s full experience
See their title, tenure and more.
Welcome back
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
New to LinkedIn? Join now
or
By clicking Continue to join or sign in, you agree to LinkedIn’s User Agreement, Privacy Policy, and Cookie Policy.
Licenses & Certifications
Patents
-
Bridging Clouds
Issued US 20150326467
A method for exchanging data across at least two different cloud service providers, comprising: receiving a virtual routing table (VRT), wherein the VRT indicates an association with a virtual local area network (VLAN) and defines neighbors for each route wherein at least one neighbor is defined for each of the two different cloud service providers, wherein the route definition creates a private transitive network between the neighbors; receiving from a first node a first message destined for a…
A method for exchanging data across at least two different cloud service providers, comprising: receiving a virtual routing table (VRT), wherein the VRT indicates an association with a virtual local area network (VLAN) and defines neighbors for each route wherein at least one neighbor is defined for each of the two different cloud service providers, wherein the route definition creates a private transitive network between the neighbors; receiving from a first node a first message destined for a second node; determining that the first message employs the route specified by the VRT; forwarding the first message to the second node; receiving from a third node a second message destined for the second node; determining that the second message does not employ the route specified by the VRT; and failing to forward the second message to the second node.
Other inventorsSee patent
Honors & Awards
-
NetApp Founder's Award
Dave Hitz, NetApp Founder
Recognition for innovation for the NetApp Data Fabric strategy, execution, and team leadership.
-
Best of Microsoft Tech-Ed 2013
Microsoft
NetApp FlexPod with Microsoft Private Cloud, Systems Management and Operations Winner
http://northamerica.msteched.com/BestOfTechEd#fbid=JFqajve4ZmK -
Government Partner of the Year Finalist
Microsoft
-
Pre-Sales Technical Specialist Partner of the Year Finalist
Microsoft
-
Server Platform Partner of the Year Winner
Microsoft
-
Management & Virtualization Partner of the Year Finalist
Micosoft
-
Private Cloud Partner of the Year Winner
Microsoft
-
Product Operations Innovation Award
NetApp Executive Team
Recognition for cross functional team leadership in the NetApp Cloud research and development of the NetApp Private Storage and Cloud ONTAP.
-
One Team Customer and Partner Award
NetApp Executive Sales Leadership
My work with NetApp customers, and partners was recognized with an annual award given out at NetApp for innovation and customer focus.
Recommendations received
14 people have recommended Lee
Join now to viewView Lee’s full profile
-
See who you know in common
-
Get introduced
-
Contact Lee directly
Other similar profiles
-
Jay Greenwald
Jay Greenwald
OnDemand Consulting for SaaS / Cloud Software
23K followersDenver Metropolitan Area
Explore more posts
-
Pradeep Madhavankutty
TransUnion • 3K followers
Just gone through the blog from MinIO on edge storage and it brings out some important perspectives for anyone building modern data or AI architectures. What I found compelling: Bandwidth continues to be exponentially more expensive than storage. Moving heavy data across the network isn't sustainable, so processing and retaining data at the edge becomes both cost-efficient and performance-driven. Their stateless, object-storage-first architecture is well aligned with how edge deployments actually operate — scalable, lightweight, and resilient even when hardware is modest or deployed in remote environments. The approach is container-friendly and fits naturally into cloud-native ecosystems, which is crucial as AI/ML workloads move closer to the edge. A few caveats worth keeping in mind: Not every use case requires full edge storage; in many scenarios, a smart cache layer could achieve similar benefits with less operational overhead. Large distributed edge deployments introduce real operational challenges — node failures, network partitions, sync issues, and recovery processes all need thoughtful design. Security becomes significantly harder at the edge, both physically and digitally, making IAM, encryption, and monitoring non-negotiable. Overall, iid the project is on the IoT, 5G, autonomous systems, or any workload where latency and bandwidth matter, this perspective on edge storage is worth reflecting on. https://lnkd.in/g_3pNzzR
11
-
Omer Khan
Google • 1K followers
The Age of Inference just got a hardware upgrade. Google Cloud just dropped two custom-silicons that will reshape cloud economics for AI and general compute: 𝗜𝗿𝗼𝗻𝘄𝗼𝗼𝗱 𝗧𝗣𝗨𝘀: This is Google's 7th-gen chip, purpose-built for AI serving and low-latency inference. For GenAI models, this means up to 𝟰𝗫 𝗯𝗲𝘁𝘁𝗲𝗿 𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝗽𝗲𝗿 𝗰𝗵𝗶𝗽 than its predecessor, scaling to thousands of chips in a single pod. 𝗔𝘅𝗶𝗼𝗻 𝗡𝟰𝗔 𝗩𝗠𝘀: The new Arm-based CPUs offers up to 𝟮𝗫 𝗯𝗲𝘁𝘁𝗲𝗿 𝗽𝗿𝗶𝗰𝗲-𝗽𝗲𝗿𝗳𝗼𝗿𝗺𝗮𝗻𝗰𝗲 𝘁𝗵𝗮𝗻 𝗰𝘂𝗿𝗿𝗲𝗻𝘁 𝘅𝟴𝟲 𝗮𝗹𝘁𝗲𝗿𝗻𝗮𝘁𝗶𝘃𝗲𝘀 for general workloads. 𝗖𝘂𝘀𝘁𝗼𝗺 𝘀𝗶𝗹𝗶𝗰𝗼𝗻 𝗶𝘀 𝘁𝗵𝗲 𝗻𝗲𝘄 𝗳𝗿𝗼𝗻𝘁𝗶𝗲𝗿 𝗳𝗼𝗿 𝗲𝗳𝗳𝗶𝗰𝗶𝗲𝗻𝗰𝘆. If your cloud costs are ballooning, these chips are designed to cut your bill while boosting your speed. 𝘋𝘰 𝘺𝘰𝘶 𝘢𝘯𝘵𝘪𝘤𝘪𝘱𝘢𝘵𝘦 𝘢 𝘧𝘶𝘵𝘶𝘳𝘦 𝘸𝘩𝘦𝘳𝘦 𝘈𝘐 𝘵𝘦𝘢𝘮𝘴 use 𝘥𝘪𝘧𝘧𝘦𝘳𝘦𝘯𝘵 𝘩𝘢𝘳𝘥𝘸𝘢𝘳𝘦𝘴 f𝘰𝘳 𝘮𝘰𝘥𝘦𝘭 𝘵𝘳𝘢𝘪𝘯𝘪𝘯𝘨 𝘷𝘦𝘳𝘴𝘶𝘴 𝘮𝘰𝘥𝘦𝘭 𝘴𝘦𝘳𝘷𝘪𝘯𝘨? #GoogleCloud #AI #CloudComputing #CustomSilicon #GenAI
2
-
Sam Naghshineh
Microsoft • 7K followers
🚀 I’m thrilled to share one of my team’s contributions in shaping the Open standard for how agents communicate their traces. With today’s announcement, Azure AI Foundry and OpenTelemetry now provide unified multi-agent observability — giving developers clear visibility into how agents plan, collaborate, and execute. This enables a unified standard into how agents emit interactions, decompose tasks, invoke tools, and coordinate with one another. 👉 What this means in practice: regardless of whether agents are built using Azure AI Foundry, Semantic Kernel, LangChain, LangGraph, or OpenAI Agents SDK, developers now gain consistent, end-to-end observability across their systems. 👉 For customers, it means faster debugging, improved safety & quality monitoring, and deeper insights into cost & performance tradeoffs. This is more than just another feature — it’s a shift in how we reason about collaboration, instrumentation, and accountability in AI systems. A huge shout-out to everyone on the team and our partners (Outshift, Cisco’s incubation engine, and beyond). Feel free to reach out if you’d like to chat about how multi-agent observability can unlock value in your systems. #Agents #OpenTelemetry #Azure #AzureAi #Observability Ankit Nagkumar QUN Priyanka Tao Mikhail Neehar Chhavi Sebastian Hanchi Shipra Yina Amanda Tina Asha For a deeper dive, checkout this excellent blog by Paul Shealy here: https://lnkd.in/gXxf-CBK
72
-
Thomas Been
16K followers
Struggling with IRSA churn during blue/green upgrades? Our Domino Blueprints guide by Principal SA Sameer Wadkar breaks down: 🔁 When to use EKS Pod Identity vs. IRSA for advanced credential propagation on AWS 🔐 How to lock down service accounts with least-privilege access ⚖️ Practical trade-offs for jobs vs. workspaces Ideal for cloud admins and security teams in regulated environments. Read the article and explore the repo for configuration examples: https://gag.gl/yzgWMW
1
-
Prateek Jain
NVIDIA • 24K followers
📣 The storage system has been reinvented for the age of #AI reasoning. Our CEO Jensen Huang sat down with NetApp CEO George Kurian to discuss how they’re bringing accelerated computing, #NVIDIA AI software, and data platforms together to solve the unstructured data challenge. NetApp AFX and the NetApp AI Data Engine enable enterprises to index, manage, and process massive volumes of data semantically across their entire infrastructure, allowing businesses to accelerate and scale their enterprise AI workloads. #NetAppINSIGHT 🔗 Watch the full conversation: https://bit.ly/3IZD03O
5
-
Binuraj Ravindran
Intel Corporation • 1K followers
Memory is one of the most expensive resources in the data center. With cloud and edge workloads pushing the limits, optimizing memory is key to lowering TCO and boosting performance. 🚀 Our new paper shows how Intel® In-Memory Analytics Accelerator (IAA) + zswap unlock efficient memory-tiering for next-gen environments. 🔍 Read more: https://lnkd.in/gnrZuASW Memory-tiering in action: With growing demands in cloud and edge computing, memory optimization is critical. 🔹 Memory-tiering—with zswap as a key technique—uses compression to expand effective memory capacity and reduce Total Cost of Ownership (TCO). The Intel® In-Memory Analytics Accelerator (IAA), integrated into Intel® Xeon® 6 processors (P-cores), accelerates compression with parallel processing. Paired with zswap, a Linux kernel feature that compresses memory pages in RAM to avoid slow disk swapping, it delivers lower latency and higher efficiency. 📑 In the paper, we dive into: ✅ Faster performance with IAA-accelerated compression ✅ Lower TCO through smarter memory utilization ✅ Scalable solutions for memory-constrained environments A big thanks to all the collaborators and mentors (Kanchana Sridhar,Rakib Al-Fahad, PhD,Vinodh Gopal,Wajdi Feghali,Ryan Saffores,Mark Adams) for their support and guidance 🙏 💬 How are you tackling memory optimization in your memory-constrained environments? Looking forward to hearing thoughts and sparking more discussions on making memory go further in modern data centers! #IntelIAA #MemoryTiering #zswap #DataCenter #CloudComputing #EdgeComputing #Intel #IntelBusiness
95
3 Comments -
Saurabh Tiwary
Google DeepMind • 26K followers
We're excited to announce new capabilities in Google Cloud #VertexAI Training designed to simplify and accelerate large-scale model development. Key highlights from our latest update: 🔹 Flexible, Self-Healing Infrastructure: Leverage fully managed, resilient Slurm environments with automated failure detection and performance-optimized checkpointing. 🔹 Cost-Effective Scheduling: Utilize Dynamic Workload Scheduler (DWS) for fixed future reservations or flexible on-demand capacity. 🔹 Integrated Frameworks: Access optimized recipes for the full model lifecycle (including SFT and DPO) and seamless integration with NVIDIA NeMo. Whether you're fine-tuning standard models or training massive custom foundational models from scratch, these new features are built to get you to production faster. Read the full announcement to see how organizations like Salesforce and AI Singapore are already leveraging these tools to elevate their models. 👇 https://lnkd.in/gVVq3WCw
185
2 Comments
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content