Still running
Fourteen of these competitions were open on 15 August. Public ranks move, and the private board is the only one that counts. Everything below is a snapshot with a date on it, and the follow-up says how the open boards closed.
Kaggle is the setting we keep coming back to, because it is one of the few places an agent loop can be wrong in public. Thousands of humans work on the same problem, the deadline does not negotiate, and the private leaderboard arrives after every decision has already been made.
What was run
Built on Humanize, we designed roughly ten distinct agent workflows and pointed all of them at live competitions. They differ in the things we think actually matter on a long run:
- how context is managed across hours of work,
- how several skills are orchestrated inside one run,
- how much of an end-to-end machine-learning engineering pipeline the agents own,
- whether a second agent reviews the first one's work, and
- whether attempts run in parallel or in sequence.
Three kinds of rank
A Kaggle result can be one of three different claims, and we count them apart.
An exact place on a closed board.
The private leaderboard, after the deadline. This is a finish.
A real score, placed afterwards.
Submitted after the competition closed and placed against the frozen final board. It is real as a score. It is not a rank, a medal, or evidence of having competed.
A position that can still move.
The public leaderboard of a competition still open. It is official, and it is a weather report.
So the honest headline is two finishes, not fourteen: Predicting Student Health Risk closed at 67th of 3,355 (top 2.00%) and ROGII Wellbore Geology at 137th of 6,125 (top 2.24%). All five top-1% results among the completed competitions are late estimates.
Every result, by kind
Each table is sortable. Click a row to pin it.
| Predicting Student Health Risk | 67 | 3,355 | 2.00% |
|---|---|---|---|
| ROGII - Wellbore Geology Prediction | 137 | 6,125 | 2.24% |
| Predicting Stellar Class | 447 | 2,816 | 15.87% |
| The 2026 NeuroGolf Championship | 482 | 2,963 | 16.27% |
The two below the top 5% are public-board ranks for competitions that had closed, where no private rank was recorded.
| NeurIPS 2025 - Google Code Golf Championship | 1 | 1,142 | 0.09% |
|---|---|---|---|
| Critical temperature of superconductors | 1 | 494 | 0.20% |
| Helios Corn Futures Climate Challenge | 1 | 153 | 0.65% |
| Electricity consumption | 1 | 150 | 0.67% |
| CMI - Detect Behavior with Sensor Data | 19 | 2,657 | 0.72% |
| Hash Code Archive - Photo Slideshow Optimization | 1 | 89 | 1.12% |
| Salary Prediction for Job Postings | 1 | 83 | 1.20% |
| Hash Code Archive - Drone Delivery | 2 | 130 | 1.54% |
| Hotel booking demand | 7 | 406 | 1.72% |
| Information Retrieval Agent 2025 | 8 | 222 | 3.60% |
| Scintillation Detector Signal Types | 4 | 98 | 4.08% |
| March Machine Learning Mania 2025 | 84 | 1,727 | 4.86% |
| March Machine Learning Mania 2022 - Men’s | 47 | 930 | 5.05% |
| Santa Gift Matching Challenge | 40 | 428 | 9.35% |
| Predicting F1 Pit Stops | 394 | 3,022 | 13.04% |
| Biohub - Cell Tracking During Development | 8 | 2,378 | 0.34% |
|---|---|---|---|
| Predicting Smartphone Addiction | 11 | 1,903 | 0.58% |
| Solar Filament Segmentation Challenge 2026 | 7 | 296 | 2.36% |
| Titanic - Machine Learning from Disaster | 262 | 10,711 | 2.45% |
| Predicting Soil Grain Size Distributions from Images | 3 | 83 | 3.61% |
| LLM Classification Finetuning | 9 | 216 | 4.17% |
| Kaggriculture | 308 | 4,581 | 6.72% |
| Spaceship Titanic | 159 | 1,635 | 9.72% |
| UMUD Challenge: Muscle Architecture in Ultrasound Data | 15 | 149 | 10.07% |
| The Pokémon Company - PTCG AI Battle Challenge Simulation | 955 | 6,829 | 13.98% |
| CUHK-X Competition Small Model Track | 31 | 199 | 15.58% |
| RSNA Knee Abnormality Detection | 300 | 1,591 | 18.86% |
| Hyperspectral Object Detection Challenge 2026 | 10 | 40 | 25.00% |
| ARC Prize 2026 - ARC-AGI-2 | 375 | 1,494 | 25.10% |
Two of the live top-3% positions need a footnote. Titanic is a getting-started practice competition. And Biohub, eighth of 2,378 that day, has since closed at 188th of 3,947: still a top-5% finish, and a smaller one.The Team Radar export behind these tables leaves out a fixed list of 12 competitions, so the counts are over the 19 completed and 14 ongoing that remain. The follow-up post counts every competition.
For comparison, Codex on GPT-5.5 at xhigh reasoning effort generally peaked at around the top-5% level across most of the same competitions. Same models, a different arrangement, and the arrangement is worth the tail of the distribution.
What we think is doing the work
Our preliminary read is that the gains come primarily from the diversity of reasoning strategies that multi-agent collaboration makes available. A single strong agent narrows early and commits. Several agents arranged to disagree keep more than one line of attack alive long enough to be tested against a validation split that has not been quietly bent to fit.
That is a hypothesis with a number attached, not a conclusion.
Further analysis is ongoing. The most controlled version of it so far is HMA on MLE-bench: two agents taking turns, against each of them alone.