# Dev Update Week 0 (example) Before the cohort starts, I am catching up on talks from the Study Group [EPF wiki](https://epf.wiki/#/eps/intro?id=study-group-content). I watched all advanced talks, took notes and asked questions about topics I wasn't sure about in the [study group discord](https://discord.gg/8RPnPGEQtJ). I read some extra materials provided in the wiki and look for something exciting to work on. There are many [projects proposed by client teams](https://github.com/eth-protocol-fellows/cohort-six/blob/master/projects/project-ideas.md), these are going to be my priority after I research couple of things I am curious about. ## Areas of research I am planning to explore 2 areas I am interested in - consensus specs and gas scaling. ## CL specs Consensus layer is specified the ethereum/consensus-specs repository which serves as the canonical source for the technical details of Ethereum's Proof-of-Stake consensus. It's a collection of documents and executable Python code that defines the rules every Ethereum client must follow to participate in the consensus process. During the initial research, I am making myself familiar with structure of the repository and its tooling. To try it out, I am writing and running my own pyspec program. To have better understanding of the actual specification, I am reading https://eth2book.info/capella, that's a great help to actually understand how CL works. ## Scaling L1 A big topic recently has been scaling L1 by increasing the gas limit. Because I am very keen on running nodes on affordable hardware, I'd like to contribute to this effort and make sure that we can run Ethereum nodes with similar hardware specs in the future. Currently, there are testnets with increased gas limit to test clients in this scenarios. [Perfnet](https://teragas.wtf/) and [Bloatnet](bloatnet.info) are dedicated to see how clients behave with increased gas limit and state growth. [Spamoor](https://github.com/ethpandaops/spamoor) is a tool for spamming networks with transactions to trigger edge case scenarios. I will connect my node to these testnets and compare its behaviour to Graphana data from Pandaops. I will mesaure how long different parts of the execution take and verify whether it reaches any limits. Using [znail](https://github.com/znailnetem/znail?tab=readme-ov-file) or a similar tool, I will simulate various network behaviour to see how client performs with limit bandwith. ## Resources Couple more cool resources I stumbled upon: - [Consensus Specifications | Technical Walkthrough]( https://www.youtube.com/watch?v=RPAf7AuLu4w) - CL design rationale https://notes.ethereum.org/s/rkhCgQteN# - Interop scaling topics https://notes.ethereum.org/@ethpandaops/berlinterop-perf