Running an AI Music Project: Week 4 Progress Report
Clafacio Lobo · Project Manager, Musinique · Humanitarians.ai
Four weeks in and the project is no longer in setup mode. The pipelines are running. The content is publishing. The research is advancing toward something submittable. And this week, for the first time, three contributors converged on a single shared creative deliverable, which is a coordination milestone worth naming.
Here is where things stand.
The Research Workstream Is Getting Serious
The most significant progress this week came from the research side of the project.
Nixon has been doing deep work on the primary paper. Five literature review gaps were identified and filled with real citations this week, that is not a small thing. A literature review with genuine gaps is a paper that does not get published. Closing those gaps moves the paper from a draft toward something credible. Nixon also built a results section template, structured and ready to populate the moment Artist.tools data arrives. The pitch to Artist.tools is drafted and ready to submit.
Shruti is working on finding annotators to validate the playlist scoring system, a critical methodological step that the paper cannot move forward without. This introduces a new external dependency: annotator availability. I flagged it as a timeline risk this week. The outreach is active and I will have a clearer picture of where it stands by next week.
The research workstream is as prepared as it can be right now. Everything that can be built without the Artist.tools data has been built. The paper is waiting on one thing, and that thing is outside our control. What is inside our control is making sure everything else is ready the moment it arrives.
The Publishing Pipeline Kept Moving
On the content side, the team published consistently this week.
Ragamalika published two articles, Before the First Note and The Listener Is the Instrument, both live on Musinique.net. Nidhi published two articles as well one on the Humanitarians.ai Substack and one on Musinique.net continuing her streak as the team’s highest-volume content contributor.
Nixon has two more articles written and scheduled for weekend publication. He is experimenting with publish timing to find better click-through rates a smart, data-driven decision that I want to track over the next few weeks to see what the data actually shows.
The publishing pipeline is healthy. The question I am carrying into next week is not whether content is being produced it clearly is but whether it is being distributed and documented consistently enough to build cumulative audience rather than just cumulative output.
A New Creative Direction — The Nana Educational Series
This is the development I am most interested in from a coordination perspective.
The Ghost Artist Nana previously focused on general music production is this week being developed as an educational content creator for children aged 3–5. Nidhi wrote original lyrics for Nana and produced an educational track on Suno. Ragamalika is building visual identity through Midjourney. Sakshi is reviewing the animations for age-appropriateness.
This is the first week all three music production contributors are working toward a single shared deliverable simultaneously. That requires a different kind of coordination than three people working in parallel on separate things. The pieces need to fit together the track, the visuals, the video and that means the dependencies between contributors are tighter than usual.
It is in progress. The pieces are being built. Whether they come together cleanly is what Week 5 is going to tell us.
What I Am Watching
Two things are carrying forward from last week that I have not fully closed.
The upload confirmation process is still inconsistent. Content is being produced and published, but the documentation layer confirming uploads to the project site, logging links centrally is still running behind where it should be. I have named this gap for three weeks now. Naming it is not fixing it. Week 5 I am tracking confirmations directly, not waiting for them to come in.
The annotator outreach for the playlist scoring validation is the newest risk on the board. Finding qualified annotators willing to do this work is not guaranteed, and the timeline for the paper depends on it. I am watching this closely.
More next week.
Clafacio Lobo is the Project Manager for Musinique, an AI music research project at Humanitarians.ai. Follow the project at musinique.net · humanitarians.ai/clafacio-lobo