Winter Hackathon 2025 – Closing Session
On May 23rd, WiBD successfully concluded its Winter Hackathon under the leadership of WiBD Global Hackathon Director Rupa Gangatirkar and Technical Advisor Stuti Patel. Rupa, an AI/ML Solution Architect and Senior Data Scientist at Siemens championed the program and served as the primary organizer and Stuti, Lead Data Scientist at Samsung provided technical guidance and coordination throughout the 8 week program. Rupa started the closing session by outlining substantial milestones achieved and future initiatives pipelined by WiBD.
A Decade of Growth and Impact
Women in Big Data has transformed dramatically since its humble beginnings in 2015. What started as 15 members gathering in a Bay Area office has evolved into a global nonprofit organization spanning six continents with over 22,000 community members across 50+ chapters worldwide.
The organization’s impressive reach includes:
- 600+ participants in mentoring programs.
- 10,000+ volunteer hours contributed in the past year.
- 8+ Hackathon cohorts, 100+ participants challenging themselves on global platforms.
- 275 learners completing certifications through DataCamp partnerships.
- 250+ training sessions conducted globally.
WiBD expands with industry specific Healthcare AI Vertical
WiBD has recently launched its AI for Healthcare vertical initiative as a global program spanning all WiBD chapters worldwide.The initiative aims to accelerate equity for women in AI and data specifically within the healthcare industry by providing targeted mentoring, networking opportunities, career development, and learning programs focused on healthcare AI applications.
Built on WiBD’s foundational services of training, mentoring, and hackathons, this vertical represents the organization’s strategic evolution toward industry-specific support, with the Winter Hackathon 2025’s ADHD detection challenge serving as an example of healthcare-focused and women health centric challenges, programs and solutions.
The Winter Hackathon Challenge 2025
This year’s competition was powered by the WiDS Worldwide 2025 Datathon Global Challenge and developed in partnership with the Ann S. Bowers Women’s Brain Health Initiative (WBHI), in collaboration with Cornell University and UC Santa Barbara. The Healthy Brain Network (HBN, Child Mind Institute), the flagship scientific initiative of the Child Mind Institute, along with the Reproducible Brain Charts project (RBC), provided essential datasets and technical support for the competition.
Hosted on the Kaggle platform, the hackathon centered on the theme “Unraveling Mysteries of the Female Brain,” with participants focusing on analyzing women’s brain health data specifically for advancing early ADHD detection and diagnostic capabilities.
Participants worked with comprehensive training datasets encompassing patients’ socio-demographic information, emotional and behavioral assessment scores, and fMRI (Functional Magnetic Resonance Imaging) data, which measures minute blood flow variations corresponding to brain activity patterns. The primary objective required teams to develop predictive models capable of determining both gender classification and ADHD diagnostic status based on the provided multimodal datasets.
Our Rocking Teams
We celebrated five incredible teams spanning four continents, multiple countries, and three WiBD Chapters alongside members of the DataCamp Donates program at our Closing Session, where they shared their journeys and learning from the 8-week challenge.
Insighters (India) – Led by India Chapter Director and Chapter Hackathon Leader Sowmya Moni, Faculty Coordinator Dr. Vijeta Khare, National Forensic Sciences University, India.
Team Leader: Aditya Kumar
Team Members: Ridyasha Ray, Adil Dyer, Ariana Ganguly
BrainStormers (Bay Area) – Led by Bay Area Chapter Co-Director Anat Zohar
Team Lead: Sriram Sridhar
Team Members: Lucy Lennemann, Niyati Bhayani, Ayah Halabi, Mansi Bhatt
DataDivas (DataCamp Donates Team)
Team Leaders: Ravi Condamoor, Srividhya Chandrasekaran
Team Members: Ann Lam, Jasmin Rafi, and Vicky Yan
DataDivas team members entered this hackathon following their successful completion of WiBD’s transformative education and professional development program through Datacamp Donates, a program designed to equip motivated candidates with essential data science competencies through the Datacamp learning platform.
Bay Brains (Bay Area) – Led by Bay Area Chapter Co-Director Anat Zohar
Team Leader: Anup Ghatage.
Team Members: Alyssa Higdon, Emily Grabowski, Winnie Cheng, Li Zhao
STLMiners (St. Louis) – Led by St Louis Chapter Director and Team Leader Usha Ramalingam.
Team Members: Zihan Zhao, Jayanth Krishnamoorthy, Justin Campbell
Team Discussions
Common Technical Challenges: All teams dealt with high-dimensional data (20K features from brain scans), extensive missing values, severe class imbalance (fewer ADHD-diagnosed females), and working with messy real-world medical datasets.
Shared Approaches: Teams universally employed different feature selection techniques, data preprocessing with imputation methods, dimensionality reduction, and advanced ML models like XGBoost and Light GBM to handle the complex multi-outcome prediction task.
Key Learning Themes:
- Transition from theoretical knowledge to practical implementation with noisy data
- Importance of addressing bias and calibrating models for medical applications
- Value of ensemble methods and cross-validation for robust predictions
- Critical role of preprocessing in handling missing healthcare data
Collaboration Insights: Teams emphasized the challenges and benefits of working across time zones (Australia, US, India), weekly team meetings, mentorship importance, and using collaborative tools like Slack, Google Colab and GitHub for coordination.
Universal Outcomes: All teams successfully submitted to Kaggle, gained hands-on experience with healthcare AI, learned advanced ML techniques, built portfolios, and developed lasting professional connections while contributing to meaningful research on ADHD diagnosis bias.
The presentations collectively demonstrated how beginners and experienced practitioners alike navigated complex healthcare data science challenges while building community and advancing gender equity in AI.
Team Testimonial
“Just wanted to say a big thank you to all of you for giving us this opportunity to participate in the datathon. We really appreciate you believing in us and providing us this opportunity. Even though we didn’t rank as high. I personally learned so much about data science through this experience only and I will always build more on everything I have learned over here. Thank you again for your guidance and support throughout. It meant a lot” – Adil Dyer
“On behalf of our team, I would like to sincerely thank you all for giving us the incredible opportunity to be part of the WiDS Datathon 2025. It was a great experience and we learned a lot and also inspired us to explore this field more. We are truly grateful to all of you for your support, guidance and the time and efforts you dedicated to helping us throughout. Although we didn’t perform as expected, we gained valuable knowledge and this experience will always motivate us to keep learning.
We look forward to using what we’ve learned and taking part in more events like this in the future. Once again, we are thankful for this opportunity and your constant support” – Aditya Kumar
Q&A Session Summary
Future Programs: Rupa shared plans for the next hackathon cohort in fall 2025 with a different scope, potentially collaborating with startups or companies for industry-specific challenges. A similar WiDS partnership hackathon is planned for Winter/Spring 2026.
Data Bias Discussion: Somya asked about inherent bias in the dataset. Team responses indicated they encountered missing data patterns and used various techniques to address bias, including location-based stratification during cross-validation, though most teams found the dataset manageable after proper preprocessing.
Community Appreciation: Multiple participants expressed gratitude for mentorship, team collaboration, and learning opportunities, with particular praise for chapter leaders like Anat Zohar and technical mentors who guided teams through challenges.
The session concluded by encouraging participants to stay connected, join WiBD’s community platforms, and participate in future initiatives.
Link to session recording is here.
