Devpost
Participate in our public hackathons
Devpost for Teams
Access your company's private hackathons
Grow your developer ecosystem and promote your platform
Drive innovation, collaboration, and retention within your organization
By use case
Blog
Insights into hackathon planning and participation
Customer stories
Inspiration from peers and other industry leaders
Planning guides
Best practices for planning online and in-person hackathons
Webinars & events
Upcoming events and on-demand recordings
Help desk
Common questions and support documentation
CS + physics at CSULB, BUILD Scholar, aiming for an MD/PhD. I build scientific software and agentic AI tools whose results can be checked.
Adapt language-dependent puzzles across languages while preserving how they are solved. Compare coordinated clues, inspect solution mechanics, and playtest the translated experience.
Turn a suspected scientific machine-learning shortcut into executable negative controls that reveal what a model learned and what it merely exploited.
Expose consequential ambiguities between speakers before fluent translation hides them, then help both people clarify and reconcile their intended meaning.
Investigate real places, compare environmental changes, and learn from measured outcomes through an evolving workspace for grounded local experiments.
Discover executable physical models from observations, compare competing explanations, and choose informative experiments to test what the world does.
Coordinate shared attention windows and route requests to qualified volunteers through explicit agreements about interruptions, availability, and response.
Return to an interrupted physical task with a deliberately confirmed checkpoint, compare what changed, and choose the next action with the context intact.
Step into the WildSide. Explore living worlds inspired by real places, meet their small inhabitants, and discover the nature hiding in everyday surroundings.
Find the first changed state in a replay, then turn that evidence into a regression test.
A Strands-powered synthetic air-traffic control agent that predicts conflicts, proposes resolutions, verifies readbacks, and fails closed when state, approval, or evidence is uncertain.
Know what arrived, what you checked, and what is still missing.
Explain the idea. Find the gap. Use it somewhere new.
Remember the promise, the condition, and what changed.
Know what remains before calling it done.
Turn an invited family visit into a personal story from approved memories.
See why smart-home automations collide, test a repair, and replay what changes.
Give residents and visitors a clear exchange across language and access needs.
Offline adaptive learning with transparent Bayesian models, reusable courses, lesson authoring, saved progress and interactive labs. Learn connections, then explain and apply them.
An offline historical heat-index replay that helps community organizers compare outdoor hours with transparent JavaScript ridge regression—not a safety forecast.