Working on https://twitter.com/ether_log

• Audited for Maker, Sommelier, TreasureDAO, Vyper and other reputable teams.

• Audited their Kiln contracts, which are used to buy back Maker's MKR tokens and burn.
• Vulnerabilities found by team and I: 1 medium, 1 low, 2 quality, 2 informational.
• Provided both audit report and statisitical assumption check report.

• Defi protocol in the Cosmos ecosystem: they offer vaults where a strageist rebalances the vault's portfolio after the strategist's trades are validated in the Sommelier blockchain and by their Ethereum contracts' rules.
• Currently has $60MM TVL.
• Report and vulnerabilities found by team and I:
  • Cellar contracts; 3 high, 9 medium, 4 low.
  • Cellar contracts and defi protocol integrations; 5 high, 8 medium, 2 low.
  • Euler integration contracts; 2 high.

• Marketplace contracts report by team and I; 4 critical, 1 high, 1 medium.

• TreasureDAO (MagicSwap).
• Frax.
• Zion (audit and design consultation).
• Vyper compiler
  • Acknowledged by Vyper team, redundant Keccak256 implementations.
  • Misc: https://github.com/vyperlang/vyper/issues?q=is%3Aissue+is%3Aclosed+author%3Aexp7l
• Based Market's Trade to Earn program
  • Solo audit; 1 ciritcal and 2 medium.

• To accelerate Darkfi ZK script developer's iteration speed, Darkfi needed a tool to generate a ZK proof's public inputs based on input witnesses, and a way to quickly prove and verify with a Darkfi ZK circuit.
• I contributed the Python bindings (1,2) for Darkfi's SDK
  • With others' contributions, it led to this demo on Twitter.

• This is a proof of concept to take Etheruem Foundation's Privacy Exploration and Scaling Group's work on their Halo2 fork to apply to Darkfi's Halo2 fork.
• The purpose is to allow Darkfi ZK circuit developer to accept additional arguments at runtime to configure their circuit.

• It is a name service that supports forward name resolution (going from name to resources) and backward name resolution (going from resources to name) in less than 200 lines of code in order to maximize auditability and security.
• A fork of DappHub's Dmap which is optimized to be as minimal as possible in implementation.

• Led and shipped automation for legal info collection and investigation resulted in $2MM saving in operational cost per year.
• Led and shipped search keyword recommendations, for amazon.com detail page. Showed $10MM+ per year profitability lift in pilot.