Delegate Proposal - 7Layer / Overclock Validator

What is the delegate’s identity and background?

Hey, my name is “7Layer” (7LayerMagik on twitter) and I will be representing the collective interests and opinions of the Overclock Validator team, which was started in 2022 and is now a midsized validator at ~1.1 million Sol stake. I’ll tell you a bit about myself along with the other two members of our Solana team.

In my “normal” life I’m currently a grad student nearing the end of a PhD program in neuroscience, but since 2021 have spent much of my free time participating in the crypto ecosystem, and in Solana in particular, where my journey into Web3 started. In 2021 I became a moderator in the Solana subreddit as well as a moderator of the Solana Tech discord, and began following the deeper core engineering and validator discussions, eventually engaging in some of these discussions myself. In the summer of 2022 I decided to start Overclock Validator with a Solana friend, Dubbelosix. Being a validator has put us in close touch with the pulse of the network and various ecosystem communities. I especially enjoy discussions on decentralization, economics, and blockchain architecture within Web3 and Solana, though I do not have a professional background in these areas. I haven’t written much formally in these areas either, but I wrote a detailed twitter thread on local fee markets that received broader attention since it was one of the first more detailed writings on this topic.

My technical cofounder, Dubbelosix (dubbel06 on twitter), entered crypto a cycle earlier than me. At Overclock he oversees devops for our Solana team as well as the Sui team. Outside of Overclock, he worked for the Web3 oracle company, Supraoracles, before moving on to work for Sovereign Labs, where he currently works full time. At Sovereign Labs, he is working on an open source SDK to enable easy development and deployment of sovereign rollups agnostic to Data Availability layers, and proof systems. While at Sovereign Labs, he has contributed to Solana light client work, benchmarking and tooling to analyze riscV ELFs to gauge prover cycles (and thus prover time) and several other core features. Outside he is also a primary technical advisor to the Solana light client team, Tinydancer. Prior to crypto, he worked in performance engineering and devops. Overall, Dubbelosix has a strong theoretical and engineering background in the realm of consensus protocols and now ZK rollup technology, while possessing good working knowledge of Defi, MEV, etc.

Recently, Shaun Colley also joined the Overclock team to lead a full time project that aims to develop a “verifying” full node client for Solana capable of running on consumer grade hardware. This project will be built upon the foundations of the Solana Go client, Radiance (found in Firedancer github). Shaun is one of the co-founders of a cyber security firm named Dark Waves InfoSec Ltd and over the years, Shaun has developed an interest in blockchain technology and security, and has earned bug bounties from various blockchains, including seven bounties between December 2021 and December 2022 from the Solana bug bounty program. He has also contributed some code to the Firedancer project (under the github handle ‘smcio’) in the form of macOS support and Gossip serialization/deserialization. On the side, Shaun has begun to explore algorithmic trading and Defi more deeply.

Has the delegate historically contributed to the Jito Network?

Overclock validator was an early stage adopter of the Jito validator client. In addition, Dubbelosix developed a quick-launch ansible playbook called “Autoclock validator” (found in Overclock-Validator github) for validators to run the Jito validator client, specifically with dedicated servers from Latitude.sh. A similar ansible playbook “Autoclock RPC” was also developed to more quickly launch RPC servers, and has garnered interest from searchers.

Has the delegate historically been involved in protocol governance?

The Overclock team is often considered one of the more engaged validator teams in the Solana ecosystem, and you can find us involved in various Solana technical discussions, mainly around decentralization and economics on the Solana discord and elsewhere. I have also been involved in similar discussions in Marinade stakepool governance and occasionally pop into discussions in various other Solana communities.

Is the delegate committed to helping the Jito Network and Solana succeed? Explain.

Yes, Overclock’s own business and our teammember’s investments are heavily dependent on the Solana network’s health and success. The Jito team have proven themselves to be one of the most important contributors to the Solana network and it is in our best interest to be fully supportive of teams and efforts that improve the chances of Solana’s long term success. We want to see that the Jito Network continues to select quality validators that improve the network, decentralization, and democratization of MEV. Our own business goals involve delivering more value back to Solana as we grow, and we are in the early stages of working toward this in engineering areas related to Solana verifiability and network robustness.

How should JitoSOL validator delegation balance performance (yield) versus decentralization goals?

Yield is important to remain competitive as a stakepool but staker yield preferences can greatly vary. Some stakers are indifferent to higher commission if they trust that the validator(s) is a good one, whereas other stakers solely care about APY and only stake with validators who are 0% commission. I think there needs to be further study of staker behavior to really know where to optimize though. Longer term I’d like to see more stakers realize that the commission “tax” they pay might return greater value to Sol if this “tax” improves the performance/decentralization of the network and also incentivizes validator contributions in other areas. It will be important that Jito broadcasts a strong opinion on validator decentralization and the benefits of higher quality validator operators given that they are one of the most respected voices in the Solana ecosystem. If there are competing stakepools that solely optimize for APY, justifying the premium paid for decentralization/quality etc will become quite important and it’s good to get the ball rolling early on establishing a strong narrative around the importance of these things.

Depending on the size of the Jito stakepool and Solana network fees, I assume that validators will be able to earn enough to run good hardware in less centralized locations, which tend to be more expensive and can be more tricky to find. This itself can serve as a good selection process to find validators who are harder working and more resourceful. I think it would also be great if there are enough geographically dispersed block engines to support more Solana network geographic spread, along with better metrics/research about the role of geographical distance from block engines and how that affects MEV returns/performance. I’m not an expert on MEV but I’m interested in understanding it more deeply and how MEV on Solana can be shaped in a way that prevents it from being parasitic to the network’s users and decentralization. Ultimately, better decentralization and reduced negative impact from certain forms of MEV makes the network a more attractive place for users and investors, which is in turn good for Jito.

What should be the priorities for JTO liquidity mining and DAO token grants?

With liquidity mining I think it’s important to be conservative here and do further research on why people choose a particular method of staking (or even not staking in some cases). Once someone begins staking in a particular way, i.e. choosing certain validators and/or stakepools, many that I have met seem to stick with the same ones out of habit and/or to avoid taxable events in the case of swapping Sol for a LST. Given this, I think it’s important to identify what people’s switching costs are and what incentives would be needed to change these habits. Related to this, it’s important to find the entry points for people into the Solana ecosystem and incentivize staking choice early before habits form. This is all a bit abstract, but something I’m interested in pondering more in the future when there’s more information at hand.

With regard to DAO grants, I think they should primarily be for Jito tooling and technical research/resources where Jito Labs may have less time to dedicate on their own. More MEV competition on Solana is good for Jito, validators, and stakers, so improving educational resources/tutorials/tooling/etc will hopefully make it easier for new searchers to join. It also seems like it might be worth brainstorming new approaches to attract engineers/traders with little to no MEV, or even crypto experience, to give it a shot, perhaps through well advertised competitions/hackathons, workshops, added gamification, etc. to pull in those with relevant engineering/trading backgrounds.

I also think that grants could/should be given for Solana R&D where Jito Labs/Foundation has identified an area with room for improvement that isn’t being covered enough by other core teams or Solana foundation grants. Also, Solana has had various fires in the past where extra help in core engineering probably would have helped, and it would be great if Jito Dao could help financially incentivize/mobilize more support in times of need. Solana’s continued success and improvement is critical for Jito on Solana.

What are your goals for the DAO as a delegate?

Our goal is to see that Jito grows in harmony with Solana itself. If chosen as a delegate, we hope that this opportunity will deepen our knowledge of MEV, decentralization, DAO governance, etc. and we want to use this power for the good of both Jito and Solana. We also like the direction that Jito is going with Stakenet and reducing stakepool trust assumptions, and we think that some of our team’s expertise can help advance these goals.

Solana Delegation address

CtZqJTwkGSCnmjQgBVPWjwXyU6dWck38xXHhDqrAC7Bc

3 Likes