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
A team project working towards AI safety applied to Google's Gemma-4-E2B model
We tried to trick Gemma into cheating - roleplay, sob stories, fake prof permission. Tested 2 languages, 4 judges, and if it acts different when "watched." Results are spicy
Jailbreaking Gemma 4 E2B models to expose safety gaps in AI lesson planning. Our goal is to highlight developer accountability and urge educators to critically review AI outputs.
Evaluate Gemma’s robustness against adversarial evasion by measuring how readily it can be induced to misclassify phishing emails as legitimate through iterative adversarial rewriting.
Evaluating the Gemma LLM on it's tutor abilities for elementary science.
A multilingual content moderation scorecard comparing Gemma-base and Gemma-IT on harm recognition, refusal calibration, and safe steering across red-teaming prompts.
Our plant toxicology and poison control assistant is a model people turn to when they suspect a poisonous plant exposure and need fast, accurate guidance.
Our project topic is a Love Island based contextual self improvement generator
Project focusing on the personal finance domain regarding budgeting, debt management, credit, investments, and more.
UVA - DS6051 Decoding LLMs Hackathon
