No funded issue found.
Check out the Issue Explorer
Be the OSS Funding you wish to see in the world.
Looking to fund some work? You can submit a new Funded Issue here .
Time left
Opened
Issue Type
Workers Auto Approve
Project Type
Time Commitment
Experience Level
Permissions
Accepted
Reserved For
Gitcoin Matching (Cerebro) - Use Case #1
gitcoinco
Python, Data
## Why
From the Milestones tracker: https://github.com/gitcoinco/data-ops/issues/41
Gitcoin's brand promise is to connect coders/contributors/hackers to funders and sponsors who are looking for skilled workers.
While we do this already, Cerebro is the foundational component that allows us to utilize the breadth of data on our contributors to match them using a variety of data points, either to relevant bounties or specific characteristics that funders are looking for. Essentially, it is an extensible tool with inputs and relevant outputs that can be used on both sides of the marketplace.
1. For example, Cerebro would aid a contributor with a Gitcoin profile, skills, and accomplishments and expose him/her to bounties that fit their experience, and/or funders and sponsors who are working on relevant projects.
## Existing Work
Currently, a barebones v1 of Cerebro has been put up here that consumes available Gitcoin data: https://github.com/gitcoinco/web/pull/5464
## Milestone 1
The first milestone will most likely involve the POC-PR that Kevin built (above) and running with that to see if it can generate positive match results for actual Gitcoin members. Our previous matching between a contributor and a bounty is purely done by tag skill matching (in the daily email).
1. So, given a Gitcoin member's profile information, can we 1. generate like-minded contributors they would be interested in, and 2. bounties that are a good match for them using this ML-algorithm?
2. Build the output of these matches into the daily email (currently IP https://github.com/gitcoinco/web/issues/6011)
3. Given these new matches, what do actual users think about them?
4. What are the measures and proxy-measures of success between how we did matching before and this new algorithm?
The output of this milestone would look like:
- [ ] productionize the algorithm
- [ ] use the algorithm to generate matches in the `new_bounty` emails that go out
- [ ] measure previous CTRs and new CTRs using the matching algorithm
- [ ] measure retention curve of email open rates
- [ ] this will give an idea whether or not the algorithm has the desired impact, and/or if we need more data to make better predictions
Setup your profile
Tell us a little about you:
Skills
No results found for [[search]] .
Type to search skills..
Bio Required
[[totalcharacter]] / 240
Are you currently looking for work?
[[ option.string ]]
Next
Setup your profile
Our tools are based on the principles of earn (💰), learn (📖), and meet (💬).
Select the ones you are interested in. You can change it later in your settings.
I'm also an organization manager looking for a great community.
Back
Next
Save
Enable your organization profile
Gitcoin products can help grow community around your brand. Create your tribe, events, and incentivize your community with bounties. Announce new and upcoming events using townsquare. Find top-quality hackers and fund them to work with you on a grant.
These are the organizations you own. If you don't see your organization here please be sure that information is public on your GitHub profile. Gitcoin will sync this information for you.
Select the products you are interested in:
Out of the box you will receive Tribes Lite for your organization. Please provide us with a contact email:
Email
Back
Save
