My journey at Consensys

I first joined Consensys as a research engineer in 2017. I had discovered Bitcoin and Ethereum only 2 years before joining. Back then, I thought that Bitcoin was the best use case of what we today call blockchain technology, but I was also intrigued about new opportunities Ethereum could unlock. 

The first project (also known as "spoke") I joined at Consensys was Linnia, a project focused on longitudinal data management, self-sovereign identity (a topic I wrote about in this book), and aimed to give users back control of their personal data. Back then, the question guiding my work was: how to reconcile privacy and transparency over decentralized and distributed systems?

After working as a Tech Lead on several projects, I decided that the next logical step of my journey at Consensys was to launch my own spoke to tackle exactly the problem of reconciling privacy and transparency. HellHound was born. HellHound is a decentralized blind computation platform. What it means is that no one, except the data provider (app user), knows the actual data being computed.

An unexpected realization for me and my co-founder Amira Bouguera (cryptographer) was that most developers didn’t actually have the mathematical or computer science background to use cryptographic libraries properly and even ended up introducing bugs when doing so. This issue of making cryptography more usable for devs has been analyzed a lot of literature, and from a security standpoint, most current DeFi smart contracts read like horror stories … 

HellHound’s goal was to provide a truly decentralized computing environment and a set of cryptographic tools to enable dapps developers to implement privacy-by-design. Unfortunately, bear market conditions and heavy internal reorganization at Consensys led to the project shutting down.

Few months later, in 2019, I went on to learn more about another protocol: Tezos. It was the first time for me working with a Proof of Stake blockchain protocol and I was attracted by the promise of formally verified smart contracts. My first thought was,“are you telling me the DAO hack could have been prevented?”

But in reality with more than 250K lines of code, getting to a fully formally verified protocol is bound to be a painfully slow endeavor. Nevertheless, this experience in the Tezos realm accentuated my belief of one point: Ethereum needed to go PoS asap or the protocol wouldn’t be relevant in the long run. 

I never thought I would actually get to play a part in that.

My story as Product lead of Besu - an Ethereum Client

I had an unconventional path to Product, which is probably what most PMs say. I started as Senior Technical Product Manager of Besu in March of 2021, after a full year of degen DeFi farming and crypto research through the 2020 lockdowns. 

My experience as an Executive Board Member of Tezos’ Research Center in 2019 was very formative, but I felt the need to get closer to the development teams again. Taking on a new role as Product Manager seemed to be a good way to achieve that and learn more about core development alongside brilliant people. 

At this point I had only one week of training - with the great Tim Beiko (ex Besu PM) - and a product with a 150K+ lines codebase to discover.

My blockchain engineering background helped me hit the ground running and my day-to-day consisted mainly of working with core devs and defining the roadmap and priorities of Besu. Working at a lower level of the Ethereum stack was both a challenging and thrilling experience for me.

The highlight of my journey with Besu was gathering with 40 core devs in Greece during the Amphora interop week in October 2021. Some of the brightest - yet humble - minds of our generation. 

Why should you care about the Merge? 

Core devs, product teams, and the broader community have been talking a lot about the Merge lately on crypto Twitter. 

Switching off PoW for PoS is a big deal as changing the consensus mechanism of a protocol that is in production without pausing it or causing a major failure qualifies as an engineering challenge. The popular comparison of “changing an airplane engine while flying” has never been more true than when it comes to the Merge on Ethereum. Especially when that plane is carrying 125.7 billion dollars of assets (TVL) in DeFi alone, 34 billion dollars worth of ETH staked on the Beacon Chain and 500K+ daily active users worldwide. This specific plane is transporting most of the emerging Web3 industry. No pressure.

Setting the right expectations as a community

The Merge isn’t a solution to all Ethereum’s issues. It’s only the next step of its evolution. You thought the Merge will reduce gas fees? Think again. 

The Merge isn’t solving directly any user experience related issues, but here are 3 very important things you can expect post-Merge at a protocol level: 

(1) Better energy efficiency with a 99.95% reduction in the network’s carbon footprint

(2) Ultrasound money with better crypto-economics and supply constraints 

(3) A seamless transition with rigorous testing since the Beacon Chain launched on December 1st 2020.

A major challenge remains: client diversity. If a client with supermajority — more than 2/3 of the staked Eth — has a problem, the entire blockchain will be affected. However, with a healthy distribution amongst clients, we can mitigate the risk of validating an incorrect, yet canonical, chain. I should also mention language diversity. About 90% of smart contracts are written in Solidity and only a minority in Vyper. Diversity at every level brings resiliency overall.

How I see the Ethereum client landscape evolving
  1. Monolithic clients such as Hyperledger Besu will evolve to become more modular, gaining both in performance and interoperability. Erigon is an Execution Client that is already embracing modularity at a core level with P2P and Web3 RPC services that can be run as components. 

  2. Beyond the first separation of concerns (Consensus layer vs Execution layer) that we are seeing in Ethereum architecture post-Merge, different roles, historically done by Eth1 clients (pre-Merge mining clients) will be externalized. An interesting example of that is the separation between block builders and block proposers (PBS). A block can be built by any independent actor (“searcher”) in a way that is more financially relevant by ordering transactions to extract maximum value from them. We call this MEV and it already represents more than 600 millions dollars since January 2020 on Ethereum.

  3. Open innovation fueled by better common goods funding will become a bigger topic than it already is. Other forums, not dedicated to “core development” but “core governance,” will probably emerge. The Client Incentive Program led by the EF and Protocols Guild are good examples of such initiatives. 

  4. Ethereum protocol as a Layer 1 might become a premium settlement layer with security primitives leveraged by other protocols offering a better user experience around fees and transaction processing. Some optimistic rollup solutions can handle ~1700 times more transactions per second than Ethereum mainnet (between 15 and 30 TPS). 

What about Hyperledger Besu?

Hyperledger Besu is an Ethereum client designed to be enterprise-friendly for both public and private permissioned network use cases.

I believe the enterprise world is starting to see that, as with internet vs intranet in the early days of the web, the real innovation and value creation is happening on the public mainnet and not on private networks. Of course various architectures can co-exist on a spectrum depending on the requirements of the addressed use case, but the endgame is mainnet.

We are observing more and more Web2 native companies making the shift and focusing their efforts on mainnet. The convergence has begun!

In the last 15 months working on Hyperledger Besu I was happy to witness and participate in the alignment of Consensys’ Besu teams towards mainnet. 

Historically, Besu has been actively participating in all Ethereum hard forks and even leading the design, implementation and testing of EIP 1559, which constitutes basically a triple halving for Ethereum. EIP 1559 was introduced in the London hardfork in 2021, thereby making Ethereum ultrasound money.

Since I joined, the number of Besu engineers at Consensys dedicated to mainnet has grown from 5 to 17, and even more resources will be directed towards researching and contributing to the implementation of Ethereum’s roadmap in the future.

I am very proud of the collective achievements of all Besu contributors and very much looking forward to this next exciting and historical chapter in the history of Ethereum.


Twitter LinkedIn

Keep up with Merge 

For regular updates and news on the Merge, visit the Consensys Merge Knowledge Base, where you’ll also find Eth2 archives with key milestones achieved and other essential resources for developers.