The fellowship ran from September 2022 through February 2023 as a four-month structured program focused exclusively on core protocol development. That means consensus and execution client engineering, base-layer research, and network specification work rather than the smart contract or decentralized application development that most Ethereum training programs cover. Twenty-three fellows received stipends. Another 13 joined under the program's open participation model, bringing the total contributor count to 36 across 20 distinct projects.

Graduates from the cohort went on to positions at Lighthouse, Teku, and Prysm (three of Ethereum's consensus-layer client implementations), as well as the Ethereum Foundation's research division, Flashbots, and Optimism. Twenty-seven core developers from active client and research teams served as mentors, and participants collectively logged more than 300 weekly development updates. Fellows also gathered in person at EPF Day during EthDenver.

Projects With Real Mainnet Consequences

Several cohort projects carried direct implications for Ethereum's live network. Fellows contributed to the consensus-layer implementation of EIP-4844, also called Proto-Danksharding, while it was still in active development. That upgrade went live on mainnet in March 2024 as part of the Cancun-Deneb hard fork. It reduced Layer 2 transaction fees by roughly 10 to 100 times by introducing temporary data chunks called blobs (stored for roughly 18 days before being pruned) to the consensus layer, replacing a more expensive mechanism known as calldata. EIP-4844 is designed as a stepping stone toward full Danksharding, which targets transaction throughput in the millions per second.

Cohort 3 work also touched EIP-4337, the account abstraction proposal that removes the requirement for users to hold ETH in order to pay network transaction fees. For anyone trying to onboard new users in regions where crypto-native infrastructure is thin, this change matters: it removes the requirement for ETH-funded accounts to pay gas, with direct relevance to onboarding users in regions with limited crypto-native infrastructure. Additional projects included research into Verkle Tries (a data structure change designed to enable fully stateless Ethereum clients and dramatically reduce the hardware requirements for running a node) and development on the Portal Network, an effort to make Ethereum accessible on devices as lightweight as smartphones.

Why the Permissionless Model Matters Globally

The EPF operates with a deliberate open-access structure. Anyone can participate alongside selected fellows, with access to the same resources, mentorship channels, and community. A portion of the stipend budget is reserved to retroactively compensate active permissionless participants, meaning a developer with no institutional backing or prior references can still earn for sustained contribution.

That design has concrete geographic implications. According to Electric Capital's 2023 developer report, 72 percent of active crypto developers are based outside North America. A separate Electric Capital report from 2024 found that Asia accounts for roughly one in three developers globally. Developer share from South Asia, Latin America, Eastern Europe, West Africa, and Southern Europe grew by a combined 20 percent since 2018. The EPF's stipend structure, negotiated individually to reflect local living costs, means a moderate monthly amount stretches further for participants in cities with lower costs than Berlin or San Francisco, making full-time participation more realistic.

The client diversity issue adds another layer of urgency. Ethereum's validator infrastructure currently runs on Geth for approximately 85 percent of validators. "Ethereum has terrible client diversity," crypto educator Cygaar noted in a January 2024 CoinDesk report, adding that "a critical issue in Geth can lead to potentially millions of ETH being destroyed." Diversifying the developer base across the teams maintaining Ethereum's other client implementations is one structural response to that risk. The program's structure suggests that engineers entering the core protocol pipeline through programs like the EPF from Africa or South Asia are positioned not just to benefit from Ethereum but to contribute to the resilience of its critical infrastructure.

The Ethereum Foundation's overall grant spending grew substantially during this period, spanning all grant categories across the ecosystem. The foundation disbursed $30 million across 397 projects in 2022 and $61.1 million across 498 projects in 2023.

What Comes Next

The Ethereum Foundation opened applications for a fourth cohort on June 1, 2023, with the window closing June 18. That cohort ran from July through November 2023 and concluded at Devconnect Istanbul. It selected 35 stipend recipients and five retroactive permissionless awards, representing an expansion over the third cohort's numbers.

Developers and technology policymakers in regions outside Western Europe and North America tracking the Ethereum Foundation's activity should distinguish the EPF from the Foundation's separate Next Billion Fellowship, which funds application-layer projects targeting financial inclusion and emerging economies. Cohort 3 Next Billion fellows included participants from Indonesia, Zambia, Japan, and Ukraine, working on projects such as tokenized artifact registries and education credential systems. The two programs address different parts of the same stack: the EPF builds the infrastructure; the Next Billion Fellowship builds on top of it. Both are relevant, but they are not interchangeable.

The Ethereum Foundation does not publish regional data for EPF participants. Geographic claims in this article reflect the program's open-access design, not confirmed enrollment figures.