The ML+X community is excited to kick off the 4th annual Machine Learning Marathon (MLM26)! Running from September to December, this semester-long hackathon offers real-world machine learning (ML) and AI projects (“Challenges“) catering to different skill levels, application areas, and ML/AI methods. This year’s challenges include:
- Building a coding agent on relatively small (<70B) open-weight models, scored on Terminal-Bench
- Transcribing 19th-century archival manuscripts with vision-language models and OCR/HTR pipelines
- Auditing a tampered neural network for planted behavior and repairing it, using mechanistic interpretability
- Predicting perceived imagery from EEG recordings
- Estimating deer populations from trail-camera imagery
- AI-emissions accounting using retrieval augmented generation (RAG)
Applications open August 19 at the end of the Challenge Showcase (4:00-5:30PM, virtual). Come to the Showcase to hear each organizer pitch their challenge and ask questions before you apply. Apply by 8/24 to be included in the first review; after that we admit on a rolling basis until seats fill. Subscribe to the ML+X Google Group for the calendar invite and the recording.
Quick Links
Who should participate?
Open to anyone affiliated with UW–Madison with an interest in applied ML/AI — undergrads, grad students, postdocs, faculty, staff, and community members. Past participants have ranged from complete beginners to AI experts.
Minimum qualifications: To set participants up for success, we ask that all participants know Python basics, machine learning fundamentals, and GitHub (enough to collaborate). If you need to get up to speed, please complete the relevant self-paced workshop(s) prior to applying: Intro Python, Collaborating with GitHub Desktop, and/or Intro to Machine Learning.
What to expect
The Marathon captures the Kaggle spirit of learning by doing. Teams of 3-6 members pick a challenge and work through it together – reading docs, searching for answers, trying things, breaking them, iterating. Each challenge lists its expected prerequisites, so you can pick one that fits your current skill level. Advisors are there when you’re truly stuck, but your teammates are your primary resource. This isn’t a guided course – expect to search a lot, struggle productively, and learn from the people next to you.
The program includes weekly in-person sprint sessions on Wednesday evenings, with async team work expected between. Participants also get an AWS SageMaker and Bedrock workshop, cloud compute credits, and ML+X Nexus as a shared knowledge base across teams and cohorts. See the full schedule for session dates and workshop topics.
Applications open 8/19. We start reviewing on 8/24 and admit on a rolling basis until seats fill. You’ll rank your top three challenges on the application, and your assignment comes with your acceptance, before you pay. Applications are per individual (not per team). If you already know who you want to work with, the application asks, and we’ll try to keep you together (assuming all teammates apply).
Questions? The FAQ covers cost, refunds, compute, and how we review applications.
“Just being surrounded by machine learning enthusiasts and professionals was really inspiring and helpful.” — Graduate student, MLM25
“One of the most impactful ML events I’ve ever been a part of.” — Academic staff, MLM25
“Collaborative work with a clearly defined goal was a nice contrast to PhD research.” — Graduate student, MLM25
“Way more interesting than my classes.” — Undergraduate student, MLM25
Help shape MLM26!
Whether you have a research problem that needs ML muscle, expertise to share, or just want to get involved — we’d love to hear from you. We’re actively looking for:
- Advisors — guide a team through the program as an ML/AI or subject-matter expert
- Presenters & demo partners — showcase a tool, method, or research application to our community
- General volunteers — help facilitate sprints, workshops, and events
All roles are welcome regardless of background. Fill out the interest form and we’ll be in touch
Thank you, sponsors!
Your support empowers Madison’s ML/AI community to tackle real‑world challenges, share hard‑won knowledge, and help one another succeed. We couldn’t do it without you.
 |
 |
 |
 |
Interested in Sponsoring Us?
Help fuel our growing community and showcase your organization’s commitment to responsible, open ML/AI. Visit the sponsorship page to learn how your organization will be represented as a sponsor of ML+X.
2026 Organizing Team
- Chris Endemann – Research Cloud Consultant | RCI
Leadership, Project Organizer, Advisor, Presenter - Ryan Bemowski – Facilitator, Data Science Hub | UW-Madison
Project Organizer - Dhananjay Bhaskar – Assistant Professor, BME | UW-Madison
Project Organizer - Yin Li – Associate Professor, BMI | UW-Madison
Presenter, Advisor - Tracy Reuter – Data Scientist | UW-Madison
Presenter
- Fraser King, Assistant Professor, AOS | UW-Madison
Project Organizer - Tejvir Mann – AI Engineer | Target Corp
Project Organizer, Advisor - Scott Prater – Digital Library Analyst, Libraries | UW-Madison
Project Organizer
- Kevin Chovanec – Professor of Practice, AI | UW Madison
Advisor - Aryan Veer Goenka – Undergraduate | UW Madison
Presenter