Currently, the MEV data tables from flashbots only support until pre-merge (15 Sep)

When you search something in the V2 search bar, the returns results based on the aliases of the data tables. e.g: if you type "erc20 transfer" you will get the result, but if you type in "transfers_ethereum.erc20" which is the exact name of the table as used in the query the search bar returns nothing. This is annoying when navigating the indexed databases as I remember the tables that I use by their proper names and not aliases. This is especially tricky for decoded functions which can have lengthy names which the alias does not reflect.

opensea v2 was not detected

Duplicate erc721, ERC721 in token_standard column in nft.trades.

Something like labels.get(...) would be amazing. Or have the full labels table available in v2 so we can join it in queries.

In Dune v2.0, a lot of abstractions now use string instead of Bytea. In SQL strings are case sensitive, so I have to format every single txhash, and address that I use because the data tables are strings now instead of bytea.

With the change from v1's bytea to strings, hashes are now case sensitive, it would therefore be a great help to have an easy built-in solution to lower case within the UI :)

The underlying asset is used as staking, it would be convenient to track this in Dune

Would be great if the Solana DB had access to dune_user_generated similar to ETH DB