Herding Squirrels

Why Your AI Pilots Keep Stalling and What to Build Instead

Louie Celiberti is the founder of E27 Technology Solutions and spent over 15 years at Guggenheim Partners leading enterprise data and AI transformation at scale. In this episode of Herding Squirrels, he returns to share why most AI pilots fail not because of the technology but because of how organizations approach the work — and what a more deliberate, architecture-first model actually looks like. If your team is drowning in AI experiments that never seem to graduate into something real, this conversation will give you a clearer picture of what’s getting in the way and how to fix it. About Louie Louie Celiberti is the founder of E27 Technology Solutions and former Managing Director and Head of Software and Data Engineering at Guggenheim Partners, where he spent over 15 years leading enterprise data, cloud, and digital transformations. He focuses on helping organizations move beyond AI pilots through a pragmatic, architecture-first approach that balances immediate business value with long-term scalability. His work centers on designing modular, vendor-agnostic platforms and investment-optimized roadmaps that allow organizations to reuse capabilities and evolve their AI and data strategy over time. Find Louie online: e27technologysolutions.io [https://e27technologysolutions.io] Episode Highlights [00:01:00] The CTO shift from driver of innovation to shepherd of innovation [00:03:25] What engineering discipline gives technologists that most business teams still lack [00:08:13] The mutual mentorship model and why it makes cross-functional AI work stick [00:12:00] Slowing down to speed up: the thin-slice approach to AI implementation [00:18:06] Meeting people where they are: how emotional intelligence shapes AI adoption in resistant organizations [00:21:12] Why collaboration isn’t optional when AI moves this fast [00:23:45] The real reason AI pilots stall: hubris, missing context, and skipping the ecosystem Key Insights * The shepherd shift: The CTO role is no longer about generating innovation from the top — it is about harnessing innovation that now originates in the business. The business has always had the context. Engineers bring the discipline to make it scalable and sustainable. (00:01:00) * Engineering discipline is a cross-functional muscle: Technologists have spent decades sitting across every business unit, learning to see how things connect. That pattern recognition — the ability to spot reusability, shared risk, and downstream impact — is something most business teams have never been forced to develop. (00:03:25) * Mutual mentorship as the change mechanism: When everyone gets a turn to share what they know, they also become more willing to listen. That dynamic creates vested interest. People who contributed to an idea will advocate for it. That is not a soft concept — it is how you move faster without losing people. (00:08:13) * Thin slices, not big bets: The instinct under pressure is to find the one transformational use case and commit. What actually works is small, multi-dimensional starting points that touch multiple perspectives at once. Narrow enough to move quickly, wide enough to represent the whole system. (00:12:00) * Skills versus talents: Skills are learnable in isolation — tools, certifications, frameworks. Talents are different. Emotional intelligence, the ability to connect dots across disconnected conversations, genuine listening — those are the things that determine whether an AI program actually takes root. (00:21:12) * The ecosystem mistake: Organizations are treating pilots like proof of concept when they should be treating them like the foundation of a capability library. Spin up five or six use cases, extract the reusable components, and you have something to build from. Learn the lesson during the pilot, not after. (00:25:01) * Progress has to be felt, not counted: The people pushing hardest for speed are usually the ones farthest from the work. The fix is not more status reports — it is framing progress in terms that connect directly to what that audience cares about, especially revenue and risk. (00:16:24) Key Quotes “Being quick is great. Being deliberate and methodical is even better — and none of those things are slow.” — Louie “The smartest people will recognize what they don’t know once they start talking to everybody.” — Louie “Skills are things you can obtain in isolation. Talents are the less tangible characteristics that allow individuals and teams to succeed.” — Louie “Assume it’s going to work. It will work. The real question now is how do you implement so that you can build other things and make it sustainable.” — Louie Resources Mentioned * E27 Technology Solutions — Louie’s firm: e27technologysolutions.io [https://e27technologysolutions.io] * Claude Code and Cursor — AI coding tools referenced for rapid prototype generation * Diffusion of Innovations curve — Referenced in discussion of AI adoption across organizational profiles (innovators, early majority, laggards) About Herding Squirrels Herding Squirrels is a podcast about modern teams and change, where we uncover the nuts and bolts of what makes teams actually work. New episodes drop every two weeks. Subscribe wherever you listen, and leave a review if this conversation was useful. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit herdingsquirrels.substack.com [https://herdingsquirrels.substack.com?utm_medium=podcast&utm_campaign=CTA_1]

Herding Squirrels Ep 23 w Stephen Szypulski

Post Body Stephen Szypulski leads post-acquisition integration at Fitch Learning, part of Fitch Group and Hearst Corporation, where the business serves more than 100,000 learners across over 100 countries. In this episode of Herding Squirrels, he walks through why so many M&A deals lose value after close, what actually breaks first when an integration plan meets reality, and why the work of integration is fundamentally about shepherding velocity without pushing the organization into shock. If you are leading change, integrating teams after an acquisition, or running an AI rollout that is starting to look like a merger, this conversation will give you a clearer way to think about people, culture, and the parts of the deal that never show up in the model. Guest Bio Stephen Szypulski is Head of Integration and Strategic Operations for Fitch Learning, part of Fitch Group and the Hearst Corporation. He leads post-acquisition integration for a global business serving more than 100,000 learners across 100+ countries, using integration as a driver of business transformation and long-term value creation. Before Fitch, Stephen held senior leadership roles at Bank of New York and Goldman Sachs, working on acquisitions including GreenSky and United Capital (RIA). He serves as an operating partner to management teams, integrating organizations, aligning commercial priorities, enabling technology, and building operating models that scale. Find Stephen online: LinkedIn [https://www.linkedin.com/in/stephenszypulski/] Episode Highlights [00:00:50] How a politics major ended up running M&A integration at Goldman Sachs and why integrations are the impetus for transformation [00:02:18] What leaders get wrong in the first 72 hours after a deal closes [00:04:23] Why human capital challenges and commercial distraction surface immediately post-close and never appear in the deal model [00:05:52] The art of velocity, or how fast you can integrate without pushing the organization into shock [00:06:18] What you can actually learn about culture from behind the diligence wall before a deal closes [00:13:32] Why culture cannot be created overnight and how phased integration earns it instead [00:21:57] Why roughly 70% of M&A activity misses its targets and what that says about how integration as a discipline has evolved [00:23:47] What AI does to the integration playbook and why humans become orchestrators of systems, not executors of steps Key Insights * Integration is structure, not a checklist. The most useful reframe in the conversation. Integration is bringing structure to chaos. It is guardrails, decision frameworks, and the work of shepherding velocity. The fifty boxes you can tick are scaffolding, not the job. (00:11:59) * The deal model never accounts for commercial distraction. Customer-facing teams slow down after close. Not because they want to. Because roles shift, products change, and people genuinely do not know who they report to or what to sell. Plan for the slowdown instead of pretending it will not happen. (00:05:27) * Culture is hard to measure from behind the diligence wall. You can read history, financials, attrition, and how an organization has handled past change. What you cannot see until after close is how a team really escalates, collaborates, and absorbs shock. Bake that uncertainty into your timeline. (00:06:18) * Velocity is the real lever. The job is to move fast enough to realize the value of the deal without pushing the organization into a state of shock that becomes disorder. Most timelines are too clean and too aggressive at the same time. (00:05:52) * Get clear about what you do not know. It is okay to enter the integration with unknowns. Build allowance into the timeline to do real diligence after close, ask good questions, and avoid making decisions that cannot be unwound. (00:17:04) * Retention is part contract and part signal. There are the structured tools, retention bonuses, role continuity, and contractual incentives. There are also the softer signals. Are you showing the people you acquired that you actually value them and the knowledge they carry. Both matter and one without the other tends to leak value. (00:18:50) * Integration work is relationship work. It touches every function and every geography, but at its core it is about how you build trust across cultures, translate between the C-suite and the frontline, and align teams around a common goal. The technical understanding matters. The relationship work is what makes it land. (00:20:41) * AI is the next integration. Bringing AI into an organization is integration work too. Systems, data, workflows, governance, and decision-making all have to come together. Going forward, the role of the integration leader becomes designing and governing those systems rather than executing each step manually, function by function. (00:24:42) Key Quotes “Integration is really about bringing structure to chaos. It is creating guardrails. It is creating decision frameworks. It is shepherding the velocity of the organization.” — Stephen “How do you move fast enough to realize the deal without pushing the organization into too much shock that it becomes disorder?” — Stephen “You have a good idea going into the deal as a part of the diligence phase. But you never really know what you get until you get to the other side.” — Stephen “It is okay to have integration unknowns that get figured out along the way.” — Stephen Resources Mentioned * BCG research on M&A failure rates (~70% of deals fall short of stated targets) * The integration-as-private-equity-operating-partner model * Stephen on LinkedIn — linkedin.com/in/stephenszypulski [https://www.linkedin.com/in/stephenszypulski/] About Herding Squirrels Herding Squirrels is a podcast about modern teams and change, where we uncover the nuts and bolts of what makes teams actually work. New episodes drop every two weeks. Subscribe wherever you listen, and leave a review if this conversation was useful. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit herdingsquirrels.substack.com [https://herdingsquirrels.substack.com?utm_medium=podcast&utm_campaign=CTA_1]

Herding Squrrels Ep 22 w Anne Marie Mills

Episode Overview What actually happens to people after an acquisition closes? Anne-Marie Mills has lived it twice — first as HR lead when Algorithmia was acquired in 2021, and again in a more recent transition spanning teams from Seattle to Eastern Africa. In this episode, Anne-Marie walks through both experiences with real honesty: the early excitement, the culture shock six months in, the attrition nobody stopped, and what she wished leadership had done differently. If you work in people operations, run a team through any kind of organizational change, or have ever wondered why the talent you acquired seems to quietly disappear, this conversation is the post-mortem you didn’t know you needed. Timestamps [00:01:03] Anne-Marie builds her LEGO model — a bridge, a flower, a slide, and a ladder out — representing the arc of the Algorithmia acquisition from excitement to attrition [00:05:05] Six months in: when the adrenaline fades, the culture difference hits, and people start asking whether they actually fit [00:06:41] Going from 55 people to an 1,800-person company — and what it feels like to realize you’re a drop in the bucket [00:09:07] What Anne-Marie would have done differently: the culture conversation that never happened and the talented engineers who left because of it [00:13:51] Communication as the first defense against attrition — and why people can’t hear anything until one question gets answered first [00:17:44] The ticking time bomb: balancing deal deadlines against the time people actually need to process a life-changing decision [00:21:12] Building space for people across 12 time zones — early morning office hours, anonymous Q&A forms, and meeting people where they ask [00:24:00] Leading through constant change: why there is no single answer and why Maslow still applies [00:27:13] The question every change leader should be asking at the six-week mark — even when everything looks fine Notable Quotes “We all understood that culture is what makes M&A successful. But unless you address the culture, you will lose the people — and with the people, the knowledge that they have.” “I sometimes feel like I have to remind leaders that you’ve hired adults. If you give people information and treat them like adults, they can help make the process better.” “It’s impossible to manage change if you don’t feel like you have something to stand on.” “What’s the message that I’m not sure landed yesterday or last month? And how can I make that message clear tomorrow?” About Anne-Marie Anne-Marie Mills is an Indianapolis-based Head of People who brings a rocket scientist’s precision to human systems. Her career began at Boeing as an aerospace engineer working on satellites and military aircraft before a Vanderbilt MBA and time at Deloitte Consulting pointed her toward HR leadership. She has led People operations for high-growth AI and FinTech startups including Algorithmia and Scratch Financial, and has managed complex people transitions for organizations ranging from 55 to over 3,000 employees. Outside of work, Anne-Marie is a passionate advocate for diversity in STEM and has visited all 7 continents and all 50 states. Find Anne-Marie on LinkedIn: linkedin.com/in/anne-marie-mills About Herding Squirrels Thank you for joining us on Herding Squirrels, where we uncover the nuts and bolts of change. New episodes drop every two weeks. Subscribe wherever you listen, and if this conversation was useful, leave us a review. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit herdingsquirrels.substack.com [https://herdingsquirrels.substack.com?utm_medium=podcast&utm_campaign=CTA_1]

Herding Squirrels Ep 21 w Frédéric Rivain

Frédéric Rivain is the CTO at Dashlane, a leading cybersecurity company — which means when AI coding tools started flooding the market, his team had more reasons to be skeptical than most. In this episode of Herding Squirrels, Frédéric walks us through how Dashlane built real AI adoption across an engineering org that had legitimate security concerns about the tools, why the hardest part of the rollout was never the technology, and what leaders get wrong when they treat AI adoption as a deployment problem instead of a people problem. If you’re leading a team through an AI transition right now, this conversation is the honest version of what that actually looks like from the inside. About Frédéric Frédéric Rivain is the Chief Technology Officer at Dashlane, a leading cybersecurity company specializing in credential management and digital identity solutions. With over two decades of experience in tech leadership, Frédéric drives innovation in security and privacy. A passionate advocate for user-centric security, he brings a unique perspective on blending convenience and robust protection for both individuals and businesses. Find Frédéric online: Dashlane Blog [https://www.dashlane.com/blog] Episode Highlights [00:01:14] Why a security-first engineering team had more skepticism about AI tools than most[00:02:04] The ambassador model: how Dashlane built grassroots buy-in before going broad[00:08:13] Using a big refactoring project as proof — what took months got done in weeks[00:09:15] Diffusion of innovation in practice: moving from enthusiasts to the early majority[00:10:31] Why Dashlane decided NOT to chase AI code generation at all costs[00:12:36] The AI guild: creating a community forum where concerns can surface safely[00:15:00] The ROI problem — why proving the dollar value of AI investment is harder than it looks[00:18:06] How AI is blurring the lines between roles and what that means for teams Key Insights * Adoption is a people problem first. The technical challenges of AI rollout are real, but the harder work is getting past hesitancy — especially in teams with legitimate, well-reasoned concerns about what the tools produce. Change management is the actual job. (00:01:40) * Start where the energy already exists. Frédéric didn’t mandate adoption from the top. He found the engineers who were already excited — the Generative AI Enthusiasts Slack channel — and gave them structure, access, and permission to experiment safely. Ambassadors create emulation. You don’t have to build enthusiasm from scratch if you look for where it already lives. (00:02:04) * Proof beats persuasion. When Dashlane used AI to help migrate a monetization system with 10,000 integration tests — a project that would have taken months manually — it finished in a few weeks. That anecdote did more to move skeptics than any internal presentation. Give people a concrete win they can see. (00:08:13) * Not all resistance is fear. Some of it is signal. When engineers pushed back on aggressive AI code generation, they weren’t wrong — they saw that generating more code faster would just create bottlenecks in review, threat modeling, and QA. Frédéric listened. They shifted focus to augmenting the whole software lifecycle, not just the code output. (00:10:31) * Community beats mandate for surfacing real concerns. The AI guild — a self-organized forum where engineers could share what’s working, what isn’t, and where they’re stuck — created a space where concerns could surface without feeling like complaints. Management was present but not dominant. (00:12:36) * The ROI conversation is coming. You can feel the productivity gain from AI. Putting dollars behind it is a different problem. Frédéric draws a parallel to the cloud transition: the cost curve and the value curve don’t line up neatly, and CFOs will eventually ask. Being ready for that conversation matters. (00:15:00) * The two skills that matter most right now. The ability to learn fast and adaptability. Not just in engineers — in every person navigating a role that looks different than it did a year ago. (00:19:12) Key Quotes "It was as much of a change management story, and the people changing their habits, more than a technical story." — Frédéric “We don’t mandate for you to use it, but you need to try. You need to play, you need to learn, you need to become fluent.” — Frédéric "One thing to call out however is that sometimes when there is resistance it's for the good reasons" “With AI, you can do ten things in parallel instead of one. But that doesn’t mean the number of things you want to do are reduced.” — Frédéric Resources Mentioned * Dashlane Blog [https://www.dashlane.com/blog] — Frédéric contributes regularly * Diffusion of Innovation (framework referenced in the conversation) * GitHub Copilot, Cursor, Claude Code — tools used in Dashlane’s AI rollout * The concept of “guilds” as community-driven knowledge-sharing forums within engineering orgs About Herding Squirrels Herding Squirrels is a podcast about the human side of change. Teams don't struggle because of bad strategy. They struggle because change is hard, people are complicated, and most organizations are better at announcing transformation than actually living it. We talk to the people figuring it out in real time. Subscribe wherever you listen. This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit herdingsquirrels.substack.com [https://herdingsquirrels.substack.com?utm_medium=podcast&utm_campaign=CTA_1]

Herding Squirrels Ep 20 w Gowri Sivaraman

Episode Overview What does it take to scale a fintech product from startup to flagship while keeping your team aligned and engaged? Gowri Sivaraman has spent 25+ years answering that question across multimillion-dollar products at companies like Intuit. In this episode, she shares the hard-won wisdom behind building teams that hold each other accountable and have fun doing it—revealing why the best teams never lose sight of their shared vision, how to remove emotion from cross-functional accountability, and what a failed basketball attempt taught her about leadership vulnerability. About Gowri Gowri Sivaraman] is a visionary technology leader with over 25 years of experience shaping multi-million-dollar products across the globe. She’s built and scaled digital consumer, small business, and fintech solutions that millions rely on every day — from AI-driven platforms to the devices we use in our pockets. Known for blending big-picture vision with hands-on execution, Gowri has a track record of turning ideas into innovations that not only engage users but also fuel significant business growth. Timestamps [00:00] Introduction and welcome[00:40] Build #1: Who Gowri is outside of work—gardening, automation, and building things that grow[02:10] Build #2: Best team experience—QuickBooks Capital’s journey from COVID startup to flagship product[04:45] Keeping your eye on metrics and KPIs as you scale[05:03] Build #3: Nightmare team scenario—when everyone runs in different directions without a shared vision[06:52] Build #4: Leading through uncertainty—adaptability, role modeling, and painting vision for the team[09:39] Staying on top of market changes: TLDR AI, conferences, and dedicating 10% of time to industry learning[12:41] Global Engineering Days at Intuit—a full week for experimentation[13:07] Building cross-functional accountability without being abrasive—using data to separate person from problem[16:09] Building fun and camaraderie: Top Gun, Top Golf, and the power of team activities[17:36] The basketball moment—how showing vulnerability inspired a team member[19:05] Build #5: Leadership advice—stay true to your north star and build a community you can lean on[20:49] Gowri’s reflection on building with LEGO bricks Notable Moments “The constant is change.” Gowri’s framework for leading through uncertainty starts with accepting that flux is the norm—and then equipping yourself and your team with the right tools and learning mindset to adapt. Separate the person from the problem. When holding cross-functional teams accountable, Gowri emphasizes using data and facts rather than finger-pointing. Lay out the evidence of what happened, and the actual person becomes secondary to solving the problem. Vulnerability builds trust. Gowri shared a story about playing basketball at a team event despite having no skills—and kept trying anyway. A team member later told her how inspiring it was to see a leader be okay with failing publicly. Sometimes the low-stakes moments create the biggest lessons. “Never take your eye out of sight.” Her best team build featured an eyeball at the top—a reminder that as teams and technology grow, you must maintain clear focus on what matters: customer outcomes and system stability. Where to Find Gowri LinkedIn: Gowri Sivaraman [https://www.linkedin.com/in/1gowri/] This is a public episode. If you would like to discuss this with other subscribers or get access to bonus episodes, visit herdingsquirrels.substack.com [https://herdingsquirrels.substack.com?utm_medium=podcast&utm_campaign=CTA_1]