Ideas Worth
Exploring.
Research. Experimentation. Design.
An environment for the next generation of curious minds.
Curiosity is the only prerequisite.
Explore CompetiLabYoung people should not merely learn established knowledge. They should learn to ask original questions, investigate problems, collaborate across disciplines, and communicate ideas that matter.
Where Research Meets Competition.
No. 01
From a vague interest to a defensible question, a careful review of what is known, and a paper written in English and matched to a real venue. Most students use data and machine learning as their instrument, with the studio's GPUs and designed modules so that the mechanics never swallow the thinking.
Two tracks: Research Foundation (4 to 8 weeks) and Research & Publication (8 to 12 weeks), with mentors who publish at NeurIPS and ICML.
About the program
No. 02
Synthetic biology the way a team lives it: a real problem, a designed biological system, a model, an experiment, and the responsibility to explain it. The road from a first idea to a high school team at the Grand Jamboree, with the modeling, Human Practices and wiki pieces prepared so the energy goes to the science.
| 01Foundations | Parts, devices, systems |
| 02Design | Problem to biological system |
| 03Model | Predict before the bench |
| 04Build | Controlled, recorded, reproducible |
| 05Engage | Human Practices |
| 06Communicate | Wiki, poster, presentation |
Creating
No. 03
Where biotechnology meets design. Students learn to look closely at living systems, imagine what they could become, and build proposals that are rigorous, beautiful and responsible. The road from a first grown sample to a team at the Biodesign Challenge Summit in New York.
About the program
Compare candidate protein variants with structure prediction, then clone only the promising ones.
Open climate and tide data feeding a model of where a living intertidal structure would survive.
Topic models over a digitized archive to show how a city argued about vaccines, pollution or a dam.
Three small models, one public dataset, and an honest study of what happens when the data has gaps.
Generative design to explore the forms a mycelium panel could take, then grow the three that make sense.
A vision model that counts and classifies colonies from a photograph of a plate, built by the team.
Every project starts with a question that could come out either way.
Mentors guide, question and critique. We provide designed modules for the mechanical parts, so building a project is like building with Lego: the bricks are prepared, and what you make with them is yours.
Students learn to say what they do not know, in their papers, models and proposals.
Work is made to be read, used or argued with by people outside the program.
We do not guarantee publication, prizes, results or admissions. We guarantee a rigorous process and candid feedback.
A short conversation about what the student cannot stop thinking about.
A mentor with research experience in a related field, and a program that fits.
Reading, questioning, designing, making. Sessions alternate with independent work.
Structured critique of drafts, models and prototypes, as many rounds as the work needs.
A paper, a project record or a proposal, shared with an audience outside the program.
Does afternoon surface temperature along walking routes predict how long students take to get home, and does shade change the relationship?
When part of a wearable device's data is missing, how does the sleep-prediction performance of lightweight machine learning models change, and at what cost in compute?
How much single-use plastic passes through one night market in an evening, where does it come from, and what would vendors and customers accept in its place?
Tell us what a student cannot stop thinking about. We will suggest where it could go.