About
I do marketing stuff to help businesses use Google's silicon - wherever and however they…
Articles by Jarrad
-
The 'Plastic Chair Protocol' – Backpacking lessons that I’m bringing to my new role in IBM Marketing.
The 'Plastic Chair Protocol' – Backpacking lessons that I’m bringing to my new role in IBM Marketing.
Another millennial on an ‘Eat Pray Love’ journey? Not quite..
13 Comments -
Why Marketers Need To Stop Chasing Trends
Why Marketers Need To Stop Chasing Trends
Research shows that Social Media Managers have a 10 times greater chance of 'burnout' than other employees. Ok, I made…
17 Comments -
3 Things You Need to Know about the New Data Breach Notification Laws In Australia
3 Things You Need to Know about the New Data Breach Notification Laws In Australia
At the intersection of innovation and technology rests a region of legal uncertainty. On a personal level as a law…
14 Comments
Activity
1K followers
-
Jarrad Swain shared thisOur engineering team is receiving strong industry recognition for optimizing TPUs for popular inference engines. Exciting stuff. SemiAnalysis, please feel free to keep the hyperbolic social posts going- seems appropriate.Jarrad Swain shared thisALERT ALERT ALERT 🚨 🚨 🚨 VLLM MAINTAINERS HAVE JUST SHOWN THAT TPUv7 CAN GET 700 tok/s/user, 56% BETTER PERFORMANCE THAN NVIDIA GB200 NVL72 THROUGH MEGAKERNEL OPTIMIZATION ON KIMI K3. As we said awhile ago, the TPU externalization of software is full steam ahead. This is ultra important to follow the progress of this.
-
Jarrad Swain reposted thisJarrad Swain reposted thisExciting news — Google Cloud was named a Leader in The Forrester Wave™: Public Cloud Platforms (Q3 2026), earning the highest score in the Current Offering category! Even better, we received the highest combined score of any vendor and the top score possible in 23 of the 30 criteria evaluated — including Vision, Innovation, AI Development, GKE / Kubernetes, Databases, and Security. As teams shift from chatbots to autonomous agents, the big takeaway is simple: you can’t run next-gen AI on fragmented infrastructure. You need a platform co-designed from the ground up — from silicon and systems to models and orchestration. What I believe is driving our leadership position: 1. Full-stack co-design: Decades of co-designing our silicon, systems, and software stacks allow customers to scale AI workloads with real enterprise efficiency. 2. Ready for the agentic era: We're continuing to evolve GKE so organizations can run autonomous agents and core workloads on a highly scalable, proven platform. 3. Grounded in live data: Our Agentic Data Cloud bridges operational databases and analytics, turning enterprise data into a real-time reasoning engine. A massive thank you to our incredible engineering, product, and infrastructure teams for their relentless innovation — and to our customers and partners who push us forward every day. Check out the full report: https://lnkd.in/g4MTV3Tv #GoogleCloud #AIInfrastructure #GKE #Kubernetes #GenerativeAI #CloudComputing
-
Jarrad Swain reposted thisJarrad Swain reposted thisWith the rapidly evolving requirements for AI models and the new demands the agentic era is placing on infrastructure, Gartner placing Google highest in execution and furthest for vision in the inaugural GartnerⓇ Magic Quadrant™ for AI Infrastructureis an incredible achievement. AI Hypercomputer is the underlying foundations to how Google trains Gemini. The composable stack delivers a unified system engineered for better performance per dollar across training, reinforcement learning, and inference. Read more here: https://lnkd.in/g3kv_5a5
-
Jarrad Swain shared thisThis is a fascinating shift for my industry, but for the non-tech or tech-adjacent folks out there, TL;DR - Companies are realizing they can't afford to use their most expensive, high-end chips (GPUs, TPUs) for every single little task. - By switching smaller background tasks to cheaper, everyday chips (like the ones made by Intel, AMD, or Google Axion), they can make AI significantly cheaper and faster. - We just launched software to make this kind of 'agent infrastructure' easier to use, which ultimately means smarter, faster AI tools for everyone at a much lower cost.Jarrad Swain shared thisI often get asked to make predictions on the future of AI Infrastructure. My short answer: While accelerators like TPUs handle core acceleration, agents need a fluid computing infrastructure with TPUs, GPUs, general purpose CPUs, storage, networking, and even specialized chips that don’t yet exist. This architectural reality is why our team moved up the general availability of GKE Agent Sandbox and introduced our new open-source project, Agent Substrate. While standard Kubernetes is built for long-running services, our framework is uniquely designed for the microsecond-scale chatter of millions of tool calls that typically overwhelm standard control planes. To support this level of scale, we require radical improvements to efficiency. Here are some samples of what we have achieved to date: Sub-Second Latency: Allocation of 300 sandboxes per second, per cluster, with 90% of allocations completing in under 200 milliseconds to eliminate cold starts. Idle Compute Reduction: Native integration with Pod Snapshots to suspend idle workloads and resume them in seconds, drastically cutting down on wasted compute. Massive Scale: We have supported 16x growth in active sandboxes on GKE since last November. This architecture handles the highest levels of agentic density and supports the orchestration of heterogeneous compute pools, including native integration of Google Axion CPUs, delivering up to 30% better price-performance than other hyperscalers. Read our full technical deep-dive into GKE Agent Sandbox and Agent Substrate here: https://lnkd.in/gbRj_qF2 And catch the full interview here: https://lnkd.in/ga-dMmkm
-
Jarrad Swain reposted thisJarrad Swain reposted thisIt’s a thrilling time for infrastructure. Reflecting on Gemini 3.5 Flash and our teams’ hard work, I’m excited about our progress. Over the next 8 days, we’ll share a daily digest, ‘8 Days of TPU 8.’ To kick off the series, I want to ground us in four shifts I am seeing across the industry and at Google. Hardware-Software Co-design: System efficiency will become pervasive in public discourse, because designing hardware and software in isolation has hit a wall of diminishing returns. Our work with Google DeepMind on silicon-software co-design allows TPU 8t to achieve nearly 3x higher compute performance per pod than previous versions. Making agentic scale economically viable: Enterprises are transitioning from large training runs to orchestrating agentic workflows, shifting architectural needs and spending to make them economically viable. Real-time, multi-turn agent interactions cause memory bottlenecks; we addressed this in TPU 8i with 3x more on-chip SRAM and 288GB of HBM to host KV Caches entirely on-silicon. And because agents can spend a large part of their runtime sitting idle, we will see a resurgence in CPU demand. Highly fluid general-purpose compute like our Axion CPUs will be vital to handle background tasks and sandboxing without compromising TCO. Designing flexible, reliable and predictive orchestrators: Agentic apps require orchestrators to transition from reactive to predictive, learning-based scheduling. Our experience with Borg and GKE highlights that manual placement heuristics fail under complex workloads. By integrating machine learning to predict task timing and arrival, we can optimize scheduling. We are implementing these strategies via the GKE Agent Sandbox and open-source Agent Substrate. Moving beyond perf/TCO: Jevons Paradox teaches us that a massive increase in the efficiency of a fundamental resource paradoxically leads to an exponential surge in its total consumption. Lowering the cost of cognition will only induce greater demand. This means that as an industry, we must look beyond TCO and engineer our systems for metrics like goodput, intelligence-per-watt and carbon efficiency to deliver increasing value at massive scale. Follow along with Google Cloud #8daysofTPU8 for more technical details and specific engineering choices powering this next frontier and catch the recording from our recent Cloud Next session on “the Future of AI Infrastructure”. https://lnkd.in/ga-dMmkm
-
Jarrad Swain shared thisJust yesterday Sundar Pichai announced that we're sending Trillium TPUs to space. And today (1 day later, approx. 8 cloud years) we're announcing that the newest generation of TPU, Ironwood, will enter GA in the coming weeks with customers like Anthropic already on board. It has been a wild ride but it's such a pleasure to be part of this team. More here: g.co/cloud/ironwood-axionJarrad Swain shared thisOur TPUs are headed to space! Inspired by our history of moonshots, from quantum computing to autonomous driving, Project Suncatcher is exploring how we could one day build scalable ML compute systems in space, harnessing more of the sun’s power (which emits more power than 100 trillion times humanity’s total electricity production). Like any moonshot, it’s going to require us to solve a lot of complex engineering challenges. Early research shows our Trillium-generation TPUs (our tensor processing units, purpose-built for AI) survived without damage when tested in a particle accelerator to simulate low-earth orbit levels of radiation. However, significant challenges still remain like thermal management and on-orbit system reliability. More testing and breakthroughs will be needed as we count down to launch two prototype satellites with Planet by early 2027, our next milestone of many. Excited for us to be a part of all the innovation happening in (this) space! Read more about it here: https://lnkd.in/enUd7trt
-
Jarrad Swain shared thisThanks for having me last night to talk about "Starting and Building a Career in Tech" at BrainStation. It's not often I'm speaking on (rather than writing the talking points for) a panel. There were so many great questions from startup founders, aspiring marketers, students, etc. Lots of fun!
-
Jarrad Swain shared thisThis week marked the launch of Google Axion Processors, a project I've been working on with Mo Farhat, Nate B., Yang Liu, Jon Masters, Rachel Richardson, Eric Gowland and many others. You folks are the best. Seeing the initial press and customer response has been so exciting, and as we wrap up #GoogleCloudNext in Vegas I can't wait to turn that anticipation into advocacy.Jarrad Swain shared this"Google is making more of its own chips, rolling out new hardware that can handle everything from YouTube advertising to big data analysis..." Read more about Axion, our custom-designed Arm-based CPU, via The Wall Street Journal ↓Exclusive | Google Expands In-House Chip Efforts in Costly AI BattleExclusive | Google Expands In-House Chip Efforts in Costly AI Battle
Jarrad Swain commented on a post
3mo
Woah, congrats Brian! That’s exciting
-
Jarrad Swain liked thisJarrad Swain liked thisCongratulations to Stephanie Buscemi on her new role as Chief Marketing Officer at Google Cloud! Stephanie has had a front-row seat to some of the biggest shifts in technology. She previously served as CMO of Salesforce, leading a global marketing team of 1,700 people, and most recently as CMO of Confluent, where she helped grow the company and show businesses how they could use data as it’s happening, rather than waiting to analyze it later. At Google Cloud, she’s taking on AI and will lead marketing as the company works to help businesses move from experimenting with AI to actually putting it to work.With women making up only about 12% of the cloud computing workforce globally, I’m thrilled to see a woman with Stephanie’s experience helping lead what comes next in AI. Stephanie, what an exciting moment to step into this role. I can’t wait to see what you do with it! The more women we see in positions of influence, the more we normalize what leadership looks like. Subscribe to the FQ Newsletter for more #InsideTrack: https://lnkd.in/gnMKEyrT
-
Jarrad Swain liked thisJarrad Swain liked thisALERT ALERT ALERT 🚨 🚨 🚨 VLLM MAINTAINERS HAVE JUST SHOWN THAT TPUv7 CAN GET 700 tok/s/user, 56% BETTER PERFORMANCE THAN NVIDIA GB200 NVL72 THROUGH MEGAKERNEL OPTIMIZATION ON KIMI K3. As we said awhile ago, the TPU externalization of software is full steam ahead. This is ultra important to follow the progress of this.
-
Jarrad Swain liked thisJarrad Swain liked thisI've spent my career on the enterprise side of technology shifts. I was at Salesforce as cloud became the default. I was at Confluent when data went from something companies stored to something they ran on. The pattern repeats: the technology arrives first, the language for it arrives late, and the companies that win are the ones who can plainly say what they do for a customer before the category even has a name. AI is in that gap right now. It also happens to be our biggest technology transformation yet. Enterprise marketing has never been louder or more crowded, and most of that noise is answering the wrong question. Enterprises are past debating whether the models are good. What I hear from customers is harder and less glamorous: Will it run in production? Will it survive our security and compliance checks? Will it hold up when ten thousand employees are on it every day? Will it break my IT budget before I see a return? That's why I'm thrilled to share that today I have joined Google Cloud as Chief Marketing Officer. Here's what tipped the scale for me: The stack, and the proof it works. Google has spent more than a decade building the answer across every layer of AI: infrastructure and TPUs, data, security, and Gemini. Enterprises don't get production-grade AI from stitching together point solutions. They get it from a stack built to work together, end to end. Nearly 90% of the Fortune 100 already use Gemini Enterprise, and 80% of Google Cloud customers use its AI products today. The opportunity now is translating that power into absolute clarity for every enterprise buyer. That's the job I'm taking. Not more volume in an already loud category. Clarity, and the discipline to keep a customer's outcomes at the center of everything we do. The people. The people made this an easy decision too. Every conversation I've had here, with engineers, sales, marketers, leaders across the business, has sharpened the same impression: deep thinkers, genuinely collaborative, customer-obsessed, with an entrepreneurial spirit you don't always find at this scale. The scale and the ambition speaks for itself. Google Cloud grew 82% year over year last quarter, with nearly 100 billion annual revenue run rate. That kind of growth doesn't happen because enterprises want another AI model to evaluate. It happens because they want AI that actually does the work: one platform, not twenty tools bolted together. My focus is making Google Cloud that platform for every enterprise, across every line of business and every industry. Not just answers. Outcomes. A huge thank you to Lorraine Twohill and Thomas Kurian for the opportunity and the trust. And to the marketing team I'm joining: I'm excited to get in the trenches with you. I'm coming in to listen first. Time to get to work.
-
Jarrad Swain liked thisJarrad Swain liked thisPretty incredible to see Google Cloud now as a Leader AND the top rated cloud in the recent Forrester Wave for Public Cloud Platforms. If you haven't used Google Cloud recently you don't know how much its evolved...Forrester Wave Public Cloud Platforms Q3 2026 report | Google Cloud BlogForrester Wave Public Cloud Platforms Q3 2026 report | Google Cloud Blog
-
Jarrad Swain liked thisJarrad Swain liked thisExciting news — Google Cloud was named a Leader in The Forrester Wave™: Public Cloud Platforms (Q3 2026), earning the highest score in the Current Offering category! Even better, we received the highest combined score of any vendor and the top score possible in 23 of the 30 criteria evaluated — including Vision, Innovation, AI Development, GKE / Kubernetes, Databases, and Security. As teams shift from chatbots to autonomous agents, the big takeaway is simple: you can’t run next-gen AI on fragmented infrastructure. You need a platform co-designed from the ground up — from silicon and systems to models and orchestration. What I believe is driving our leadership position: 1. Full-stack co-design: Decades of co-designing our silicon, systems, and software stacks allow customers to scale AI workloads with real enterprise efficiency. 2. Ready for the agentic era: We're continuing to evolve GKE so organizations can run autonomous agents and core workloads on a highly scalable, proven platform. 3. Grounded in live data: Our Agentic Data Cloud bridges operational databases and analytics, turning enterprise data into a real-time reasoning engine. A massive thank you to our incredible engineering, product, and infrastructure teams for their relentless innovation — and to our customers and partners who push us forward every day. Check out the full report: https://lnkd.in/g4MTV3Tv #GoogleCloud #AIInfrastructure #GKE #Kubernetes #GenerativeAI #CloudComputing
-
Jarrad Swain liked thisJarrad Swain liked thisEnterprise AI is facing a token-cost reckoning. As AI shifts from chat to agents, a single request can trigger hundreds or even thousands of machine-to-machine interactions, driving 50–100x more inference transactions. Google Cloud now processes more than three quadrillion tokens every month. 7x more than a year ago. So how can enterprises keep AI infrastructure both scalable and economically sustainable? In new research developed with Google Cloud, Futurum Research examines the economics behind agentic AI, from cost per successful run to prompt caching, model routing, batching, and infrastructure choice. And at The Six Five Summit: AI Unleashed 2026, Daniel Newman and Google Cloud Security’s Mark Lohmeyer explore the infrastructure side of the equation, including a striking gap: 90% of enterprises want to deploy agentic AI, but only 17% believe their infrastructure is ready. 📄 Download The Token Cost Reckoning Arrives in the CFO’s Office: https://lnkd.in/gEAmeB6S ▶️ Watch the full Six Five Summit conversation: https://lnkd.in/gJXtxjdA
Experience & Education
Licenses & Certifications
Volunteer Experience
Honors & Awards
-
Collegiate Scholarship
Bond University
-
Premier's Award for Academic Excellence
NSW Government
Recommendations received
2 people have recommended Jarrad
Join now to viewOther similar profiles
Explore more posts
Explore top content on LinkedIn
Find curated posts and insights for relevant topics all in one place.
View top content