YearnV3 - Request For Budget

[jmonteer]

Motivation

This effort will cover the top layer (yVaults). The design will be done to fit in the above's diagram bigger picture.

https://medium.com/iearn/yearnv3-motivation-and-design-107840cb4844

Objective

Build the next version of yearn vaults.

This includes V3's core code and smart modules that replicate current functionality (using multi-sigs for permissioned functions).

Main objective of V3's core code is to get closer to a protocol (vs. current product) that will allow to build current product offering on top.

Core code will handle the most basic logic in a robust way:

• Funds distribution
• Share management (deposits, withdraws, price per share, ...)
• Role management
• Strategy management (add strategies, remove strategies, ...)

Monitoring, tx execution, compounding, debt allocation, will be done at product level and implemented in Smart Modules or Periphery.

It will also allow for third parties to develop their own products with a different set of compromises: more centralized, more risky, less liquid, less expensive...

Wishlist:

• easy to write strategies
• protocol >> product (zero maintenance)
• modular architecture
• immutable core
• gas efficient

Current status

Up to date, we've been working in the design process and initial iterations (ported from Vyper V2 and made changes to reach 80% of feature completion) This is where we are sitting today.

• Tech Spec:
  • not written
• Code:
  • Core code is close to feature complete
  • Smart modules and periphery work has not been started
  • Written in Solidity (80% completion), ported from V2 vault's Vyper
• Tests:
  • Tests have been written in TypeScript using Hardhat as framework
  • E2E tests: some progress made
  • Unit tests: some progress made
• Team:
  • 3 external contributors have been helping
    • 2 devs @ 16.5k/mo
    • 1 PM @ 2-3k/mo (Coordinape)
  • 3 Doers

Deliverables

• Technical Spec
• Code:
  • Vault (Core code)
  • VaultFactory
  • Smart modules replicating v2 functionality
• Tests in ApeWorx
• Support for code deployment and initial set up

Team

Accountable Doer: jmonteer

• jmonteer for product and protocol dev
• Panda: Protocol dev doer
• Bounty program for external contributors
• Stakeholders:
  • Security
  • Strategists
  • B2B
• Klim will help coordinate audits

Timings

• Phase 0: (mid June)
  • comparison of Vyper vs Solidity
  • close collab agreement with ApeWorx
  • write yearn's v3 vision
  • write tech spech (similar to https://github.com/yearn/yearn-vaults/blob/master/SPECIFICATION.md)
  • complete rewrite in Vyper (if approved during last benchmark)
  • change tests suite to use ApeWorx and Python
• Phase 1: alpha version: (expected mid July)
  • funds come from treasury exclusively (1-5m)
  • strategists build different strategies: some are possible today, some won't be possible today
  • several ways to recover funds along the way will be allowed (using permisioned functions like sweep, ...)
• Phase 2: beta version: (expected mid Aug)
  • guestlist enabled by a marketing stunt (NFTs, veYFI holders, ...)
  • deposit limit is capped around 20m
  • several ways to recover funds along the way will be allowed (using permisioned functions like sweep, ...)
• Phase 3: stable version: (expected mid Sept)
  • open version
  • request migrations from v2: partnership with wido
  • core code is immutable and won't have guardrails

Budget

Budget is expected to be distributed to external contributors through a bounty program. Expected to be paid at the beginning of the Phase. Phase 0 and Phase 1: $45k Phase 2: $20k Phase 3: $15k

(+ auditing costs)

Send budget to 0x370D98fB4E5A0fC6cb92b256c3d5883f5cbDa5fD Please, fund with a small amount of ETH too.

[jmonteer]

https://etherscan.io/tx/0xa1c1635bdd4786361b3cb39322bc78ba4511986fb517f1709ef04c7b3bfa9244

Expenses were paid according to bounty levels: dev-lvl-0: 300 DAI dev-lvl-1: 700 DAI dev-lvl-2: 1400 DAI dev-lvl-3: 2000 DAI