Back to the Lab

Challenges

Pick a challenge, join the team, ship a contribution.

2Open now
28 000CP in play
7Projects
14 challenges shown
MLMedical Imaging AI
12 000CP POOL

Mammography Classification

Classify a mammogram (normal / benign / malignant) or BI-RADS with an open-source model fine-tuned on public datasets, evaluated on AUC.

0 contributions this week · 2 active
7%
Contribute
MLMedical Imaging AI
16 000CP POOL

Mammography Segmentation

Delineate lesions on a mammogram at pixel level (segmentation mask), with an open-source model evaluated on Dice/IoU.

0 contributions this week · 2 active
0%
Contribute
CodeLeaderboard
8 000CP POOL

Sandbox Proposal Integration

Turn the leaderboard into a two-way platform. Anyone can propose their own health project, and the community stars the best ones. Star milestones pay the author, and an admin can promote a proposal into an official project.

Delivered and evaluated · CP distributed
100%
See results
CodeLeaderboard
4 000CP POOL

Code Space Workflow

Shape the contributor workspace for code challenges. Each contributor gets a personal kanban and a dedicated branch (or their own repo). They work the full scope in parallel and trigger a single evaluation of their whole delivery once their board is done.

Delivered and evaluated · CP distributed
100%
See results
CodeLeaderboard
12 000CP POOL

Conversation Collaboration Patterns

Reward the collaboration that happens outside the code. Managers define contribution signals ("proposed an idea", "helped a peer") with a fixed CP value. A daily AI pass reads the challenge's Slack channel, detects those signals and credits authors automatically, with an auditable justification for every reward.

Delivered and evaluated · CP distributed
100%
See results
CodeLeaderboard
3 000CP POOL

ML Collaboration Workflow

Design how ML contributors work together on datasets, models and API packaging. Each step is scored and paid live from a draining pool. Reusing someone else's dataset or model sends a share of the points back to its original author.

Delivered and evaluated · CP distributed
0%
See results
CodeLeaderboard
4 000CP POOL

Determinist Evaluator Grid

Build the evaluation backbone: structured JSON grids (criteria, sub-criteria, scoring anchors) and the AI agent that scores submissions against them. The goal is reproducibility. The same delivery must get the same score, verified by repeated test runs before a grid goes live.

Delivered and evaluated · CP distributed
100%
See results
CodeLeaderboard
8 000CP POOL

Leaderboard PoC

Lay the foundations of the platform: challenges with CP pools, contributor profiles, an append-only reward ledger and the public leaderboard. The goal is to prove that contributions can be tracked, scored and rewarded end to end.

Delivered and evaluated · CP distributed
100%
See results
CodeMedical Imaging AI
12 000CP POOL

Mammography

Detect breast cancer from mammograms

Delivered and evaluated · CP distributed
63%
+6
See results
CodeInjury Prediction
12 000CP POOL

PoC: Injury Prediction in Tennis

Predict sports injuries in Tennis based on player's data

Delivered and evaluated · CP distributed
100%
See results
MyTwin Lab Community Management
6 000CP POOL
Delivered and evaluated · CP distributed
81%
Read more
MyTwin 3D
10 000CP POOL
Delivered and evaluated · CP distributed
32%
Read more
CodeMyTwin Avatar
4 000CP POOL

Live Face Surgery Preview

See nose, lips, chin and jawline changes in real time on your own moving face.

Delivered and evaluated · CP distributed
0%
See results
CodeMyTwin Avatar
2 000CP POOL

MyTwin Avatar

A realistic 3D preview of your own body, from a few smartphone photos. A few photos taken with an ordinary phone are enough to fit a 3D body model.

Delivered and evaluated · CP distributed
0%
See results
Got a project of your own?
Create your sandbox