What I Learned Reorganising a 45-Person Team at the Ethereum Foundation

In mid-2025, the Ethereum Foundation restructured its Ecosystem Development function. I led that restructuring.

This is what I learned about building an organisation inside a decentralised ecosystem - what worked, what didn't, and why the ultimate measure of success is whether what you build can thrive without you.

The Starting Point

When I took on EcoDev, the function was primarily oriented around grants. Applications came in, they were evaluated, funding went out. It was reactive by design. The logic made sense in an earlier era: Ethereum needed to fund public goods, and the EF had the treasury to do it.

But the world had changed. Enterprises were showing up and finding no front door. Founders were building on Ethereum without anyone helping them navigate fundraising, positioning, or investor introductions. The app ecosystem - DeFi protocols, identity platforms, social apps - had almost no relationship with the Foundation at all. Developers had learning resources but no coordinated growth strategy. Communities existed between conferences but had no permanent infrastructure.

The grants model wasn't failing. It was just solving yesterday's problem.

The Design Principles

I came into the Ethereum Foundation from a commercial background - I'd built and exited e-commerce businesses, turned around underperforming divisions at Morgan McKinley as their youngest-ever director, and doubled revenue year-on-year scaling a remote sales organisation at a Berlin edtech startup. That turned out to be an unusual advantage.

Most people in foundation or protocol roles come from research, engineering, or community backgrounds. Those perspectives are essential, but they tend to produce organisations optimised for technical correctness rather than market responsiveness. Coming from environments where you either shipped and sold or you went under gave me a different set of instincts: obsession with speed of execution, a bias toward direct engagement over process, and a deep discomfort with building teams that couldn't articulate the strategic rationale for their own existence.

That tension - between the EF's mission-driven, long-horizon culture and a founder's instinct for velocity and market feedback - shaped every design decision in the restructuring. I didn't try to make the EF operate like a startup. But I did insist that every team we built could answer the question: what would change in the ecosystem if this team didn't exist?

In hindsight, that question turned out to be more than a management heuristic. It's a version of what the Foundation has since formalised as the Only-EF Rule: focus on critical work that has no other natural home. If someone else can do it, they should. If no one else will, we step in - temporarily.

I restructured EcoDev around a few core convictions.

First: teams are hypotheses, not permanent structures. In an ecosystem as fast-moving as crypto, organisational design has to be fluid. A team that makes sense in Q1 might need to be dissolved or merged by Q3. This isn't instability - it's intellectual honesty about where the ecosystem actually needs support. We built EcoDev with the expectation that its shape would change, and that change would be a feature, not a failure.

Second: presence beats funding. The EF had historically supported the ecosystem through capital allocation. That's necessary but insufficient. What enterprises, founders, and app teams consistently told us they wanted was someone to talk to. A point of contact. Someone who understood their problems and could route them to the right people. So we shifted the centre of gravity from grants distribution to direct engagement.

Third: design for graduation. The goal was never to make EcoDev the permanent gatekeeper of ecosystem relationships. Every programme we built - community hubs, founder matching, the app discovery surface - was designed to reach a point where it could sustain itself independently. We wanted to create infrastructure the ecosystem would own, not make the ecosystem dependent on our continued involvement. The EF's long-term success should be measured by how much of what it builds no longer needs it.

What We Built

We went from a grants-oriented function to a cross-functional organisation spanning twelve teams across six workstreams. Here's how each one came together and why.

Enterprise Relations - Building the Front Door

Twelve months before the restructuring, a Fortune 500 company interested in Ethereum had no landing page, no point of contact, and no clear entry point. The world's second-largest blockchain network had no enterprise front door.

We built one. I hired a five-person enterprise team distributed across the Americas, Asia, Europe, and the Middle East. The goal was global coverage within a single quarter, and we hit it.

We launched institutions.ethereum.org as a public resource, which generated organic coverage from Yahoo Finance, Bankless, and multiple industry outlets within the first week. More importantly, we started showing up in places the EF had never been: inside bank headquarters, at Sibos Frankfurt, Money2020 Riyadh, and in closed-door forums with senior regulators in Korea.

One month we presented to a major UK asset manager. The same month, we presented at the House of Commons. That range - from institutional investors to parliamentarians - was the whole point.

The design decision that mattered most: we didn't embed enterprise people inside existing teams. We created Enterprise Relations as a standalone function with its own mandate, its own rhythm, and its own accountability. Enterprise engagement requires a fundamentally different cadence than developer relations or community building. Mixing them dilutes both.

By the end of the first six months, the team had spoken to hundreds of enterprises and, for the first time, made Ethereum consistently accessible to institutions exploring the network. We built an institutional privacy task force, began developing a Certified Solutions Provider Program, and announced an Enterprise Advisor Network for Q2 2026.

What matters most about this work isn't that the EF is doing it - it's that it's now proven enough to exist independently. The demand is real, the playbook works, and the function has its own gravity. That's the signal that it's ready to graduate.

Founder Success - Getting in the Room

The conventional approach to founder support in crypto is hackathons and accelerator programmes. We tried something different.

Instead of running open funnels, we worked hands-on with early-stage teams on fundraising, positioning, distribution, and business development. Not advice from a distance. We helped rewrite pitch decks. We made introductions to the right investors. We were present in the room during pitches.

This was a deliberate bet against volume. We rejected the idea of high-volume intake because founder support adds most value downstream of quality filtering, not by processing applications. The result: we helped unlock lead investors and term sheets for teams building critical Ethereum infrastructure - projects like Glacis, Upshot, Null, Index Network, and onchain.city, each chosen because they were building self-sovereign tools the ecosystem needed. We supported the migration of projects from competing chains to Ethereum with a combined TVL of over $500 million - projects that moved because they needed Ethereum's security and credible neutrality, not because we offered incentives.

The key learning: we built a dynamic where introductions from EcoDev carried weight because we had credibility in filtering for quality and alignment. That's not something you get from running a programme. You get it from consistently backing the right teams with real support.

We also launched a founder clinic with real-life mentor-to-founder matching - first in Buenos Aires, now scaling globally. And we learned that classic funnel models like generic hackathons have diminishing returns. The strongest outcomes came from partners with real skin in the game.

App Relations - Repairing a Trust Gap

This was perhaps the most sensitive workstream. The relationship between the Ethereum Foundation and the application ecosystem - DeFi protocols, social apps, identity platforms - had deteriorated. App teams didn't know who to talk to at the EF. They felt unsupported and, in some cases, actively ignored.

Fixing this required face time, consistency, and visible support. I personally reached out to every major Ethereum app team and set up intro calls. Weeks of back-to-back calls. We created direct group chats between app teams and EcoDev. We hosted internal EF calls where app teams demoed real products, exposing protocol researchers to app-layer realities for the first time.

One of the best hiring decisions I made was bringing in DeFi-native people for this team rather than generalist relationship managers. Having people who actually understood the protocols massively upgraded the credibility of conversations and the quality of support, especially for institutional bridging use cases.

We shipped ethereum.org/apps - an official, neutral app discovery surface that replaced the vacuum left by DappRadar's decline. We created an app-focused event at SuperRare gallery in New York, featuring thirteen Ethereum apps across DeFi, identity, and social. It was the first time many app builders felt visible in an Ethereum-native context.

The app discovery surface is a good example of infrastructure that should outlast any particular team. It's neutral, it's public, and it serves the ecosystem regardless of who maintains it. That's the kind of thing worth building.

Digital Studio - From Website Team to Content Engine

The ethereum.org team had been doing solid work maintaining the website, but it was underutilised. We rebranded it as Digital Studio and repositioned it as a full content and storytelling operation.

The AI integration here was transformative. The team built workflows that save an estimated 100+ hours per week - roughly two to three full-time equivalents of output, automated. They replaced approximately 2,500 hours per year of outsourced translation work with automated AI pipelines. One person now runs a full video production operation that simply wouldn't have existed without AI tooling.

The output speaks for itself: the OFFSTAGE series produced over forty founder interviews. Amplification videos for the Lean Ethereum and dAI teams each crossed one million views. When quantum computing concerns hit mainstream media, we had a response video out in 72 hours. The Ethereum 10-year anniversary campaign drove over 700,000 pageviews and became the most widely shared EF campaign to date.

But the most strategically important numbers are these: ethereum.org sends approximately 100,000 users per month to wallet providers through the wallet explorer, and over 18,000 developers to SpeedRunEthereum. That's ecosystem infrastructure, not content marketing. Those pipelines run regardless of who's overseeing them.

Analytics & AI - Correcting the Narrative

We formalised a dedicated Analytics & AI team in August 2025. By January, they had already changed the public narrative about Ethereum's developer ecosystem.

The team audited Electric Capital's Developer Dataset and found a massive volume of previously uncounted repositories. The corrected number: 66,359 active Ethereum developers - more than double the next largest ecosystem. The narrative that Ethereum was losing developers was wrong. We proved it with data.

They built a custom binary classifier for GitHub repos with a 0.85 F1 score that evaluates full codebases via LLMs rather than just READMEs. They deployed daily ecosystem scrapers tracking around 300 accounts for narrative analysis and competitive intelligence. They automated the weekly fundraising report. They launched Stratus - a tool that unifies views across multiple EF team CRMs, creating the first searchable EF network.

The learning here was that centralised automation doesn't scale as well as embedded expertise. We're now pivoting to an "AI champions" model: people embedded in each team who understand their own workflows. Education first, tools second.

Ethereum Everywhere - Community Infrastructure

Between conferences, builders in most cities had no permanent place to connect, learn, or collaborate. We launched a Community Hub programme: physical spaces with regular programming, locally led and designed for self-sustainability.

The model is lean by design: local leadership, a strong focus on self-sustainability. These aren't EF-funded coworking spaces. They're convergence points for events, learning, startups, builders, and new users - and critically, they're designed to run without us.

Active hubs now operate in San Francisco, London, Berlin, Lagos, Dubai, and Hong Kong. A new hub launches every 30 days. We partnered with Gitcoin for GG24 and funded ten meetup organisers across Latin America for a full year, keeping post-Devconnect momentum alive.

The community hubs are probably the clearest example of the graduation principle in action. Each one is locally owned, locally operated, and locally sustained. The EF provided the catalyst, but the hubs belong to their communities.

What I Got Wrong

A few things I'd do differently.

I underestimated how much pure marketing doesn't compound. Early on, we invested in visibility efforts that generated impressions but didn't create durable value. The things that compounded were formats that combined genuine impact with visible proof of impact - founder success stories, app launches, data corrections. Purely private wins don't compound, and purely promotional efforts don't either. The highest leverage sits at the intersection.

The initial high-touch approach was necessary but I should have planned the transition earlier. We spent weeks in back-to-back calls with every app team, every founder, every enterprise lead. That was essential to build trust and understand pain points. But I didn't build the handoff mechanisms early enough, which meant the teams risked becoming dependent on relationships with specific people rather than sustainable institutional connections. In an ecosystem that values decentralisation, single points of dependency - even well-intentioned ones - are a vulnerability.

I hired generalists in a few roles where I needed specialists from the start. The DeFi-native hires on the app relations team were a turning point. They changed the quality of every conversation overnight. I should have applied that principle more broadly and earlier: in crypto infrastructure, domain credibility isn't a nice-to-have. It's the difference between being taken seriously and being tolerated.

The Numbers

In 150 days, the restructured EcoDev built ecosystem infrastructure now operating across 74+ countries:

The Principle

The conventional wisdom in large organisations is that restructuring is disruptive and should be minimised. I think the opposite is true in fast-moving ecosystems.

The crypto industry changes faster than most organisational designs can accommodate. If your team structures are permanent, they will eventually optimise for self-preservation rather than impact. The restructuring I led at the Ethereum Foundation was built on the premise that every team is a hypothesis about what the ecosystem needs - and hypotheses should be tested, validated, and sometimes retired.

We shifted the Ecosystem Support Program from reactive intake to intentional, priority-led funding with RFP and Wishlist rounds. We reduced total grant spend while making funding more targeted, so impact per dollar went up, not down. We moved from managing existing funding sources to building long-term economic resilience - mapping what sustainable funding actually requires to sustain Ethereum for the next hundred years.

The restructured EcoDev wasn't built for permanence. It was built with graduation in mind. The mandate was to create functions, prove they work, and prepare them to stand on their own - so that the ecosystem's capacity grows while its dependency on any single organisation shrinks.

The Foundation's role has always been to steward Ethereum into maturity. Mature infrastructure doesn't need a permanent steward. It needs good foundations - and then the freedom to grow beyond them.

That's what I think more organisations in this industry need: not better strategy documents, but the willingness to build things that are designed to outlast their builders. Build what matters. Build it to last. And build it so it doesn't need you.

← Back to all writing