Thesis is a cryptocurrency venture production studio. We build products and protocols to support and protect
- Self-sovereignty
- Privacy as a human right
- Opt-in culture
We've built a portfolio of cryptocurrency projects, backed by major investors in the space.
- Fold, an online and in-store payments and rewards app originally launched in 2015.
- Keep, a data confidentiality platform for Ethereum. Keep enables off-chain stakers to provide provable sMPC services to smart contracts on the EVM.
- tBTC, a bonded Bitcoin sidechain built on Keep and Ethereum.
We'd like to add Zcash to that list, further developing and supporting the chain and ecosystem as a major dev fee recipient.
Our vision for Zcash is the "privacy layer for the Internet of Money".
Today, two public chains are economically relevant: Bitcoin and Ethereum. While both have made efforts to address privacy, they've suffered from each project's respective direction and split focus.
Privacy on Bitcoin has focused on incremental improvements. Small-set mixers and obfuscation on Lightning are state-of-the-art, but don't offer robust privacy guarantees.
Unlike Bitcoin, Ethereum makes privacy experimentation by developers fairly accessible. Unfortunately, on-chain privacy constructs are expensive and compete with other applications of the chain for block space, and no solution has gained significant traction.
Zcash is well-positioned to act as the privacy layer for these chains. By supporting user-defined assets and encouraging bridges from other ecosystems, Zcash can maintain relevance and drive privacy in a multi-chain world.
Over the years, we've developed core competencies applicable to the Zcash ecosystem.
-
Engineering — our experienced team of software engineers have built popular applications and infrastructure in and outside the cryptocurrency space. In addition to our portfolio, we've built multi-currency wallets and traditional and cryptocurrency payment infrastructure, contributed widely to open-source, and shipped to prod at Silicon Valley powerhouses.
-
Design — our design team works across the spectrum. For live examples of our portfolio of visual, interaction, and UX design, check out ln.pizza, Fold, My First Bitcoin Pizza, and the visual identities of Keep and tBTC.
-
Dev/Ops — our engineering team has strong operational experience. We follow a declarative approach to infrastructure-as-code, automating testing and integration heavily for a good developer experience.
In our experience, this is often overlooked in the cryptocurrency space. Strong tooling and cross-client testing means easier onboarding for developers looking to build on Zcash.
-
Legal — we've invested to make legal a core competency, involving legal feedback early to inform product decisions, while gating guidance to control costs.
-
Business development — rather than rely heavily on marketing, we've focused on the network we've built in the space, growing relationships and strategic partnerships.
Thesis is a full-stack cryptocurrency studio, but we do have some weaknesses.
- Research. Our team implements cryptographic papers and does minor translational research, but we prefer to partner with researchers rather than produce novel research in-house.
- Marketing. While we run growth campaigns for the products we launch and maintain, we're still formalizing our marketing efforts. We wouldn't be a strong pick to directly market Zcash, instead growing the ecosystem through products, partnerships, and infrastructure improvements.
Rather than proposing a roadmap which depends on other recipients' priorities or the price of ZEC, we'll propose a workstream that plays to our strengths and can be adjusted based on market conditions and progress across development teams.
We'd like to prioritize
-
connecting Zcash to other chains
-
ensuring ZEC is accessible from censorship-resistant DEXs
-
moving toward the orderly deprecation of
zcashd
To that end, we propose the following projects, the detailed planning of which we'll defer.
tZEC is a Zcash sidechain anchored to Ethereum, based on our work on tBTC. The initial version of tZEC will make use of Zcash's Flyclient improvement, the recent addition of Blake2 in Ethereum, and t-addresses. The reliance on t-addresses will be removed in a subsequent version.
Why?
- ZEC can be accessible via Ethereum-based decentralized exchanges, providing resilience to delisting on centralized exchanges.
- Zcash grant recipients can use Ethereum to hedge their earnings using collateral-based stablecoins like MakerDAO.
- This move will continue to align the historically friendly Ethereum and Zcash communities, maintaining Ethereum dApp developer interest in Zcash.
The Zcash Golang client will be the third Zcash client implementation, based
on btcd
.
Why?
- More independent implementations mean more resilience to bugs and attacks on the chain.
- Building a Go client will involve contributions back to the Golang crypto community, raising awareness of Zcash and helping grow its open-source presence.
- Having a Go implementation will enable easier collaboration with Ethereum and Cosmos developers, whose primary clients are written in Go.
- Finally, completion and a successful test period on mainnet will allow the safe deprecation of the C++ codebase. Ditching the remnants of the legacy Bitcoin codebase will make bigger protocol improvements easier across clients
In close partnership with the ECC and ZF, we'd like to bring user-defined assets (UDA) to Zcash.
Why?
- User-defined assets will afford stablecoins, Bitcoin, and other digital assets the same privacy guarantees as ZEC.
- Many assets on the same network will increase Zcash utilization and grow the anonymity set.
The second version of tZEC will focus on a bi-directional bridge between Ethereum-based assets and UDA on Zcash, and remove use of t-addresses.
Why?
- Bitcoin on Zcash will be possible via tBTC
- Decentralized and fiat-backed stablecoins will become usable on Zcash
- Obsoleting t-addresses will increase shielded adoption
The Ethereum / Zcash bridge can be generalized and extended to operate on the Cosmos network, lowering the cost of future interoperability work and ensuring that Zcash is the privacy choice for the budding Cosmos ecosystem.
Why?
- As Cosmos launches IBC, each supported chain will increase the number of assets that can be shielded on Zcash
To devote significant resources and attention to Zcash over other opportunities, we'd like to receive 25% of the dev fee.
In exchange, we'll organize a new business unit, focused on Zcash development and following the workstream above, as well as priorities agreed upon with other dev fee recipients and the Zcash Foundation, when appropriate.
We'll target a 20% minimum margin on our work, ensuring we have enough profit to smoothly respond to changes in the market, as well as upside as we drive value to the ecosystem. This margin is in line with what a typical digital agency or software development shop might make, versus the higher margin of specialized engineering services typically expected in the cryptocurrency space. We believe this low margin will be offset by the security of an ongoing allocation and future coin appreciation.
For the life of our allocation, we'll scale the team up and down based on market conditions and our outlook, ease of hiring and onboarding, and development priorities.
As the first proposal of this sort that we're aware of, we welcome feedback from the Zcash Foundation and the public.
Some diligence, however, might not be appropriate in public forums due to financial, competitive, or legal sensitivity. In those situations, we welcome the Zcash Foundation to do diligence for the entire community, and are prepared to share further operating details they believe are relevant.