Gurkan Yilmaz

Between 2003 and 2006, Amazon went through a period of concentrated process invention. During those years, Jeff Bezos drove the development of a set of foundational operating practices that became the spine of Amazon’s culture and execution. When people think of Amazon, they often think of groundbreaking services like 1-day delivery or innovative products like the Kindle, Alexa, or AWS. However, those services and products were only able to achieve such great success because of the processes that were behind their development. Of the Amazon processes that were developed between 2003 and 2006, you may have heard of these especially famous ones: → “Working backwards” and the PR/FAQ → Single-threaded teams/leaders → Bar Raiser Hiring → Input/output metric reviews → The six-page narrative The reason that Jeff and the leadership team were able to develop so many powerful mechanisms that are still in use to this day is that the process was iterative and grounded in a willingness to experiment and learn. These processes came from trying to solve specific organizational pain points, not theoretical problems. As process innovators, Amazon learned from traditional management science while also standing on the shoulders of other great companies and management thinkers. They adapted concepts to their unique scale and pace. It worked, and it stuck. The important thing is that these weren’t hypothetical or academic ideas. They were forged under pressure and constantly iterated until they reached a lasting form. Amazon’s true genius isn’t how they have delivered a single product or service, it is the creation of systems that allow them to consistently produce innovative products at scale. After writing “Working Backwards,” Colin and I realized that we could have called the book “The Invention Machine,” because that is truly what Amazon is. If you want to read that book, you can find it here: https://lnkd.in/gzJ4qb45

142

2 Comments

Search and news advertising ex TAC revenue grew 21% and 20% in constant currency driven by volume and rate growth across Edge and Bing. Bing and Edge took share for the fourth consecutive year. https://lnkd.in/gyDinYME

39

Excited to share my conversation with Marc Baselga and Ben Erez on the Supra Insider podcast. We went deep on one of the biggest shifts happening in product leadership right now: why many experienced PMs and product execs are returning to individual contributor (IC) roles in the AI era, and why that’s not a step back, but a strategic leap forward. Other topics we discussed: -- Why the leverage equation has changed—and how managing AI agents now rivals managing people -- The critical importance of hands-on credibility in AI-native companies -- Why great PMs in the future will blend product, design, analytics, and frontend execution -- What it takes to deliver real customer outcomes (not just launch features) -- How to evaluate breakout companies—and what founders often get wrong when hiring product leaders I also shared my thoughts on: -- Joining the right company: mission, trajectory, talent density, and founder courage -- Crushing it as a “super IC”: the power of speed, ownership, and infectious energy -- Transitioning from product leader to GM or founder with the right mindset If you’re navigating your next chapter in product (or just trying to stay sharp in the AI-native world), I hope you’ll find this useful. Episode links: Spotify: https://lnkd.in/gK48PFWK Apple: https://lnkd.in/gmT265BY YouTube: https://lnkd.in/gtr4RAKd Substack: https://lnkd.in/gM38FSWc

96

10 Comments

In 2026, we’ve reached a point where every "chat" with AI has a tangible footprint: the Water Bottle Rule. On average, a brief back-and-forth of 20 to 50 prompts "consumes" about 500ml of water. As the AI processes your request, servers generate intense heat, and data centers use evaporative cooling, essentially "sweating" out fresh water into the atmosphere to keep the hardware from melting. When we treat AI like an infinite, free resource, we overlook the fact that each prompt is a tiny withdrawal from a local water table or power grid. This is why being mindful of how we prompt isn’t just about "efficiency", it’s about digital stewardship. By optimizing our workflows, grouping our tasks, and writing clearer, more intentional prompts instead of "trial-and-error" chatting, we aren't just saving time; we’re literally saving water and energy. #mindfulprompting

2

Our Q3 FY25 results are in and Klaviyo delivered another outstanding quarter. Demand is strong. Our innovation is relentlessly accelerating. AI is increasing our value creation for customers. Revenue grew 32% year-over-year to $311M, with a non-GAAP operating margin of 14.5%. Fewer than ten public software companies are growing above 30% at a $1B+ run rate while also hitting the Rule of 40 — and Klaviyo is one of them. At Investor Day, we outlined  three engines fueling our long-term growth, and our Q3 results show clearly that we’re delivering on all three: 🔹 Multi-product expansion: We added 7,000 new customers this quarter, bringing us to 183,000 total, up 17% year-over-year. More than half of our ARR now comes from multi-product customers. And Service is off to a fast start – just six weeks in, adoption is growing faster than SMS was at this stage. 🔹 International acceleration: We expanded SMS to 22 countries, added four new languages, and launched WhatsApp. Together, EMEA and APAC revenue grew 43% year-over-year, with EMEA gaining even more speed from Q2. 🔹 Mid-market & enterprise momentum: Our customer base spending $50K+ in ARR grew 36% year-over-year, adding a record 272 net new accounts. These results underscore the strength of our model and the power of AI as a force multiplier for our customers and for Klaviyo. We’re defining what it means to be the B2C CRM for the AI era — built on data, powered by intelligence, and driven by an incredible team. The opportunity ahead of us is massive — and we’re just getting started. https://lnkd.in/g5EMZrc3

324

2 Comments

Excited to share how Wayfair is leveraging OpenAI to transform their e-commerce operations! Rather than treating generative AI as just a side experiment, Wayfair has successfully embedded OpenAI models into their core internal workflows—and the results speak for themselves. Here is how they are driving measurable impact at scale: Catalog Accuracy & SEO Boost: Wayfair transitioned from using expensive, bespoke models to a scalable, tag-agnostic AI architecture. They have already corrected 2.5 million product tags across over a million highly-visible products, leading to a significant increase in SEO performance, clicks, and customer discovery. Meet "Wilma": To handle complex supplier support requests, Wayfair introduced agentic AI workflows. Wilma automates ticket triage, fills in missing context, and suggests draft responses. The result? They are now automating 41,000 tickets per month (up to 70% in some workflows!), drastically reducing manual work and turnaround times. Empowering the Workforce: Beyond operational workflows, Wayfair has deployed over 1,200 ChatGPT Enterprise seats to help their 12,000-person team problem-solve, experiment, and move faster every day. It’s a fantastic masterclass in using AI to not only augment human expertise but also build a better, more trusted shopping journey for customers. Read the full case study here: https://lnkd.in/gyUPsV7s

29

Everyone is racing to build bigger AI models. But the real bottleneck in AI today isn’t compute. It’s operations at scale. As GPU density increases, cooling shifts to liquid and immersion, fabrics move to ultra-low latency, and multi-tenant AI factories become the norm, one thing becomes clear: AI doesn’t fail because of silicon. It fails because of operational gaps. “Operational excellence isn’t defined by automation alone. It’s defined by disciplined execution, resilient teams, and processes designed for failure before it happens.” At the scale of thousands of GPUs, AI infrastructure becomes less about hardware and more about precision in design, in monitoring, in incident response and in leadership. This is where competitive advantage is actually created. The organisations that will lead the next decade of AI won’t just ship faster chips or larger clusters. They will master repeatable execution, operational clarity, fault tolerant design and the teams that perform under pressure. Automation gives us speed. Hardware gives us power. Operational discipline gives us trust, and that is what allows AI to scale globally. As the year closes, I am excited by where AI is heading, and even more focused on how we build it. Because the future of AI won’t be decided in product demos alone. It will be decided in data centers that never go dark. Wishing friends, colleagues a happy new year with lots of excitement!

24

1 Comment