I agree, reddit and github discussion shows enough interest to move to funding required.
Cutting wallet sync time
Who am I?
Developer of one of very few miners in crypto-currency world that did not recycle BTC CPU mining stack written 7 years ago. I am a freelance developer so this is my day job :). I program mainly in C and C++.
What?
Improvements that, taken together, will at very least double sync speed. Technical details are discussed on github.
This is subdivided into two separate goals.
Milestone one
Around 1.5 weeks of work - $2250 - 125XMR
Create and integrate into Monero a special version of the crypto-ops.c that uses amd64-64-24k for the bignum operations. This library will only be used for syncing wallet and only on amd64 builds (Note that both Intel and AMD cpus use amd64 instruction set. ARM cpus are not included in this proposal)
Milestone two
Around 1.5 weeks of work - $2250 - 125XMR
Improve multi-threading and data-feeding in the wallet itself. This will involve creating separate threads to decouple the fetching from processing in order to enable full cpu utilisation as long as the bandwidth (network, disk or DBMS) is not maxed out.
What is not included?
I will work on CLI wallet only right now. If there are any breakages in GUI wallet (might happen especially around milestone two), they will need to be fixed by the GUI wallet developer or as an extra goal at a later date.