Codex Update May 14th - May 20th
The Codex team wrapped up a successful demo at the Logos off-site and now aims to prepare the demo for the Q2 public testnet release potentially coinciding with EthCC.
An internal devnet has been running for the past few weeks and has been used to test the latest version of Codex and can be accessed using the Codex Testnet Starter documentation.
Development is currently broken into three distinct teams:
- Client, Testing, and Infrastructure
- Marketplace
- Research
The following are their team updates.
Client, Testing and Infrastructure
Completed
- Packaged self-contained LevelDB Nim project so no dependencies on system packages lifted from existing wrapper and then wrapped it in a datastore implementation
- Preliminary tests shows increased performance is as expected
- Resolved issue with dependency bump and broken dependency chain and now stuck with serialization issue - bug was related to resolving to wrong deserializer and work on nim-serde and nim-RPC
Ongoing
- Started work on LevelDB migration; Discord bot, reward system and chain inspection broken
- Resolving failing integration tests re: expiry of the request in storage
- Working on issue related to inability to see amount of tokens sent in block explorer
- Deploying faucet web app for testnet tokens (ETH & TST)
- Working on Codex release workflow
- Trying to use DST’s k8s cluster and have not been able to use it yet
- Migrate dist test cluster from Digital Ocean to Hetzner and deploy different mointoring tools (e.g., Victoria Metrics) - parametrics logs replacing elastic search with similar API and CLI scrollingto be added to UI later on
Marketplace
Completed
- Helped with some work related to LevelDB integration
- Added custom errors PR in nim-ethers that required attention
Ongoing
- Writing documentation around some of the Solidity code ahead of audit
- Coordinating audit with SC team (Vac) and looping in Security; still deciding on 2nd security audit company
Research
2024 R&D Goals
1. Proving system and aggregation improvements (folding or lookups)
2. Aggregator/validator design
3. DHT improvements
4. Tokenomics and incentive design
5. Bandwidth incentives
6. Dynamic data (appendable data)
7. Encryption
Completed
- Sync w/ Client team about
circom-compat - Studying bulletproofs with the goal to try to use them in proof recursion, on the “wrong” curve - see writeup here
- Updated thinking about proof recursion
- Wrote a high-level overview explaining all the moving parts of a ZK proofs (different files and steps and their meaning)
Ongoing
- Proposal for tracking proofs (more precisely, the multiset commitments in there) has a scaling problem, and started thinking about an alternative motivated by Plookup
- Looking more into Winograd-type FFT