Melissa Evers

5G upgrades loading… ⚡️ Nokia picks Intel Xeon 6 with E-cores for its newest 5G core network solutions, delivering up to: ✅ 150% more performance ✅ 60% less power consumed ✅ 60% smaller footprint Read more. intel.ly/4eyZfZS

Building #GenAI apps with @OPEAdev? This article covers orchestration, microservices, and real-time data using Amazon Bedrock and OpenSearch: intel.ly/3HNb7Lc

Check out my commencement speech from yesterday - University of Oregon, Computer and Data Science linkedin.com/pulse/real-wor… via @LinkedIn

the Milky Way is not only flying at 600 km/s, but also “flapping its wings” like a butterfly🦋

A really long line of #opensource hackers outside the @UN office. We’ve been cooking this for the last few months and glad to see this finally happening today!

It’s not just about what you build, it’s who you build it with. Welcome @BudEcosystem to the @OPEAdev partner ecosystem. Learn more: intel.ly/45WkLoL

@intel @IntelBusiness

Listen to @jzemlin 's keynote at #IntelVision2025 as well as my thoughts on #opensource #AI's future. "The sleeper super power to drive exponential growth in Open Source AI?" Governance linkedin.com/pulse/sleeper-… via @LinkedIn @linuxfoundation #KubeCon2025

Our new Open Edge Platform empowers you to bring edge AI to market faster and more efficiently. Visit us at #KubeCon booth S331 for a demo!

Intel Unveils Leadership AI and Networking Solutions with Xeon 6 Processors #IAmIntel bit.ly/4hSISrS

In partnership with @intel, @UCBerkeley just announced a new #AI Center of Excellence focused on optimizing vLLM performance on Intel hardware platforms. What does this mean for the future of AI? Learn more: 👉intel.ly/3ZzJpH7

📣 Your voice matters! Take part in @Intel's Open Source Community Survey and help shape the future of #opensource at Intel. You could win one of 4 $250 cash gift cards! intel.ly/422zJHX 🙏 #opensource #intel

Good summary of efforts across numerous teams to empower #devs with #performance and #opensource .

Thank you, @SecRaimondo, for your confidence in @Intel and highlighting our commitment to boosting domestic chip production, with the support of #CHIPSAct funding. Watch the full @CNBCOvertime interview: x.com/CNBCOvertime/s…

🎉 How can #OPEA help you? OPEA marks its 1.0 release with 16 end-to-end examples of how GenAI can help solve specific enterprise needs. Learn more and join the growing community of 40+ partners: intel.com/content/www/us…

🎧 Give a listen to the latest episode of "How AI Happens" for Intel VP @melisevers's take on what makes a healthy open source community, GenAI challenges and opportunities, and what she's most excited about where tech is heading next! #OpenSource #GenAI intel.ly/3zJZW2n

@arungupta's interview with the @semaphoreci podcast is live! He discusses @intel's open source work using Open Platform for Enterprise AI (@OPEAdev) to support the widespread adoption of AI within the enterprise. Tune in to the full episode here 👇🏼 intel.ly/3MX6FJc

#IntelGaudi 3 AI accelerators help you leverage #GenAI tools to scale without limits and develop without constraints. Let the AI acceleration begin. #StartswithIntel

Founder mode vs Bureaucrat mode Many of you have heard about Founder mode, the idea that there are times when you have to make real decisions, overrule people, lead via conviction not consensus, etc. Before this, Ben Horowitz wrote is the excellent essay on Wartime CEOs vs Peacetime CEOs preceding “founder mode” by over a decade. But let’s be honest with ourselves as we read all of this. It’s very aspirational to work somewhere where you see leadership from across the company working towards high-conviction success. But it’s rare. What’s more common? Bureaucrat mode. This is what happens when companies get big, scaled, and successful. Here’s Bureaucrat Mode: - create committees for every decision - make sure every meeting has pre-meetings to build the papers/deck for the meeting - end every meeting by expanding the scope of the project - no one ever owns a decision, so make sure there's complete consensus. Create more meetings if needed - punish anyone showing initiative - discipline anyone who moves quickly without complete consensus - require detailed status reports before any progress - create complex approval workflows for trivial tasks - celebrate vanity metrics and milestones - reward people based on "impact" based on how may people are working on your projects - ask legal, brand, compliance to approve everything no matter how small - talk endlessly about downside risk I’m sure as you read this that this is starting to sound familiar. For all of us that have worked in scaled organizations — as Uber was when I left, at 20,000+ people — a lot of these are actually flavors of “best practices” that people purposely implement. Note that I didn’t put OKRs, QBRs, brand editorial guidelines, unnecessary legal/compliance review, etc on here but they probably should be! It’s easy to beat up on these ideas We can read the above list and laugh (and cry a little too) but of course they fundamentally are the result of good intentions. After all, we’re forming committees to facilitate communication when very complex initiatives like products are getting launched. There’s just a lot of details, and a lot of tradeoffs, and not everyone agrees. This is the good interpretation of this. And to extend these root causes further, there are a few ways that the road to hell is paved with good intentions: - collaboration: Let’s get everyone to work together :) - consensus: The need to address everyone’s concerns, to avoid mistakes, so that everyone is supportive of the decision - inclusiveness: Making sure everyone’s opinions are heard - stability: The core business is the gravy train, and why endanger it? - empowerment/delegation: You don’t want to step on toes! Let’s trust the various teams do their work! - accountability: We have KPIs and goals and playbook, and def not worth distracting ourselves I’m sure these all sound familiar — we’ve all said words like this! Taken independently, of course these are positive+helpful cultural values, and when you take these and then implement them in the form of processes/committees/etc., they can be great. But when industrialized on a massive scale as large tech companies do, it becomes hard to get anything done. Of course, this is where startups have a huge advantage over over large companies. When you have 2-3 people, there’s no consensus that needs to be reached over weeks of meetings — all the information is already held within peoples’ heads, as they work from the same room. You can move incredibly fast and just focus on output, because there are less social relationships to manage. It’s fine to disagree, because it either takes a moment to sort out, or you can just try stuff out, and undo if it doesn’t work. But all of this bureaucracy is not just driven by good intention. What makes this Bureaucrat mode and not Collaboration mode is that often these mechanisms are hijacked by people who forget why the mechanisms exist in the first place, and instead use the machine to drive their own careers. Self-replicating bureaucrats If you create an organization where “impact” is measured by how much your team is outputting — and thus, it correlates with the size of your team — then you are going to create a massive incentive to pitch all sorts of large scale projects that require hiring. If people see that other people getting promoted requires them to manage people, so that their responsibilities and scope are vast, rather than the success of their output — well, you are going to creative an incentive to hire a ton of folks. If big visible projects (“Project XYZ!”) end up being what’s required to drive internal visibility, and thus promotions, small impactful things will be ignored and big grandstanding projects will end up being encouraged. Committees will be formed for reasons other than building consensus. This creates the phenomenon of self-replicating bureaucrats: If winners hire winners, and losers hire losers, what do bureaucrats hire? More bureaucrats of course. The reason is that companies that highly prize consensus, process, etc., will inevitably hire the people who are good at executing against this set of constraints. This continues and continues, until the moment the company is required to actually move nimbly to face off against an entrepreneurial new startup (example: car companies versus Tesla) or a big technology trend occurs (example: AI and Europe). Because there’s so much that’s unknowable about these situations, and so much of what’s required is just to try things and learn things fall apart. The highly consensus-driven, collaborative organization that has become staffed with self-replicating bureaucrats end up not being able to bureaucrat themselves out of the situation. The cycle of life This phenomenon is so ubiquitous that it’s almost a cycle of life within tech. - A new, nimble startup with an aggressive new founder(s) emerges - To scale, it hires well-intentioned, competent managers - It wins the market (woohoo!) and IPOs - Later, bureaucrats who are attracted to peacetime (and brand, and stability) sneak into the company. They have shiny resumes - The entrepreneurial people quit, or leave, and can’t deal with the new processes. The bureaucrats take over. The founder either checks out, or retires - The company is in Bureaucrat Mode - A new, nimble startup then emerges… Without this cycle of life, the tech industry would not exist. I saw this first hand at Uber, which was respected as the fastest-moving big company led by an aggressive founder, and eventually things got bogged down as it grew. Very hard to counteract, even with a company where “moving fast” was part of the core DNA. In tech, at least we have a cycle of life where new startups can take over. In Europe however… :)