Hi everyone,
Following the Code & Capital Hackathon, we are partnering with RubberDuck on the Code & Capital × RubberDuck AI Code Audit Challenge.
What this is
Your AI coding agent can generate code fast.
But does it actually understand your codebase?
As projects grow, the hard part is often not writing more code. It is knowing whether the code is correct, whether a change breaks hidden dependencies, whether a security path is real, and whether the agent is guessing from context or reasoning from the actual system.
RubberDuck gives AI coding agents deeper codebase intelligence: call chains, data flows, dependencies, definitions, security paths, and change impact.
This challenge is designed to test what RubberDuck surfaces in real Code & Capital community codebases that regular AI agents miss, would likely miss, or cannot prove as well.
How it works
- Install RubberDuck.
- Connect a codebase you are authorized to use.
- Run UC-01: Understand Your Code.
- Run UC-02: Codebase Audit / Security Audit.
- Submit your transcript, one-page findings writeup, codebase description, and 60-second video.
Optional add-ons include bug localization, code review, change impact, and feature planning.
You can use a hackathon project, startup product codebase, side project, internal tool, or open source repository. Python, TypeScript, and JavaScript codebases are preferred.
Submit your audit
- Submission form forms.gle/WuwvioB4QtK7uKcj9
- Participant brief What Did Your AI Agent Miss? — Read the brief →
- Live setup support Join the RubberDuck support Slack
- RubberDuck docs rubberduck.com/#docs
Deadline
Tuesday, May 19, 2026 at 11:59 PM PT
Prize
One winner will receive a 4-hour working session with Marco Marinucci or Jordan Greene, General Partners at Essentia Ventures.
This is direct investor/operator time to pressure-test what you are building: product strategy, technical roadmap, codebase, AI workflow, fundraising narrative, market thesis, or any major strategic question.
What makes a strong submission
The best submissions are specific.
Show what RubberDuck found, where it appeared in the codebase, why it mattered, and what your regular AI agent missed or could not prove.
Selected submissions may be featured in a Code & Capital × RubberDuck recap or research note on what AI agents miss in real codebases.
Please do not submit secrets, credentials, customer data, private source code, or anything you are not authorized to share.
Best,
Code & Capital