[Proposal] Numerai Youtube Tutorial Series

Council of Elders
#1

Hi everyone! Long-time lurker and occasional shit poster. I’ve been experimenting with some Numerai Youtube videos and I am pretty happy with the results so far. So, I’ve decided to make a proposal so I can get some feedback from the community, and see if the community wants more proposals around content creation.

  1. Proposal // A summary of the project you’re proposing and who is going to be working on it
    My aim for these videos was to do shorter how-to / tutorial style videos for the tournament (& Hopefully signals in the future if everything goes well!). I’ve posted the engagement from the first 4 videos of my series and am looking to do around 4 more (I try to keep the videos ~10-15 mins long).

The remaining topics I was looking to cover were:
• Training a Numer.ai model in the cloud (For those having memory issues with their local setup)
• Staking, Scoring & Diagnostics
• Automatically submitting predictions with NumerCLI
• Buying/staking submissions via NumerBay.

StudyM8_YTanalytics
StudyM8_YTanalytics1252×625 50.2 KB

  1. Timeline // When could you start working on the project, if approved; And how long would it take?
    I can start immediately, producing a video biweekly. Can have the 4 remaining videos done by Jan 31st.

  2. Best case outcome // What does success look like?
    Best case outcome is more people are exposed to Numer.ai through my videos and join the tournament, signals, or community.

  3. Worst case outcome // What could go wrong? The more thought that goes into this, the better.
    Nobody watches the videos // low engagement and new users. If the community does not like my style/content in the videos, perhaps harm Numerai’s reputation.

  4. Success criteria // How can the success of this project be objectively measured and evaluated?
    Was hoping to rely on YouTube/Twitch analytics for the most part here. I am hoping that with better tagging, titling, editing, etc. I can improve engagement, but hard to ballpark a target (Maybe 2.5K views cumulative views by Jan 31st? Open to suggestions here)

  5. Funding required, if any.
    So far, it’s been taking me a couples days to research/code/debug/record/edit each video. So I’d ballpark ~ 8 hrs of work going forward per video (~12 NMR using the rate from previous proposals).

Here’s my channel, StudyM8 - YouTube . Appreciate any comments/critiques/feedback in advance :slight_smile: !

8 Likes
#2

This is my third time trying to get into Numerai Turnament.
Granted, the first time I knew too little about ML and python.
The second time, I still did not know enough about ML.
And now I did some kaggle and actually know how enough to get around a little.
I watched all the videos and find them so extremely accessible. Peter is doing a great job explaining the concepts of the numerai and explains some good techniques on how to handle the data.
There are other videos out there, like from the numerai team itself. But they all are fragmented into these office hours and what not.
So as a person who is trying to get into Numerai, I think it would be a great Idea to fund him to keep making these videos.

4 Likes
#3

@malding I’ve dropped your proposal link into the CoE chat channel on rocket chat. They should have someone assigned to it and back to you shortly. Good luck! I can see this being really helpful to the newer users.

2 Likes
#4

Thank you @objectscience ! Much appreciated.

1 Like
#5

These look really cool. I’m in favor of funding these and kind of fits in with the new process of paying after each deliverable is posted.

Just as a quick temperature check, what are general thoughts: should CoE fund this proposal?

  • Yes
  • No
  • Other (explain in comments)

0 voters

3 Likes
#6

Added to our proposals board: CoE proposals · GitHub

3 Likes
#7

Looks like this definitely has the necessary support for funding. CoE multisig queued 40 NMR to your Numerai wallet address for retroactive funding on the previous 4 videos and will follow up with funding after you post each subsequent one :slight_smile:

6 Likes
#8

Congrats @malding! looking forward to seeing this grow!

1 Like
#9

Thanks @objectscience. Shout out to everyone who voted and assisted with the proposal, I appreciate the support. Excited to get the next couple videos out! :slight_smile:

6 Likes
#10

Congrats @malding :slight_smile: I’m also looking forward to the next videos :tada:

2 Likes
[Proposal] Bounty for high quality data science posts
#11

Hey everyone! Next video is up here.

Sorry for the delay, needed to learn some Dask/ test out a bunch of Cloud providers for this one. 8hrs or 13.85 NMR with a current 28.89 NMR price.

I’m good with using the same wallet as last time, but let me know if I need to post it here again.

Hope you all like it and again any feedback is greatly appreciated. Thanks!

4 Likes
#12

Awesome work! Keep them coming!

1 Like
#13

How did you get merch? I want that shirt bad

#14

Hey did you never receive funds from the last video? Also, do you plan or releasing more Numerai tutorial content or do you feel the set is complete?

#15

Hey! No I didn’t receive them, and yes still plan on doing more videos, just got a bit sidetracked with work, and they take longer than I had initially thought when I made the proposal :sweat_smile:

1 Like
#16

hit me up in rocketchat so i can get your address details that you want the NMR sent too

#17

Going to have 14.6 NMR queued up for you since we missed this one updated to todays pricing of 27.39. Sorry about that and next time hit one of us in the CoE up in rocket DMs or the CoE rocketchat channel if we miss something in the future.

1 Like