Junior Agentic Developer

What you will do:

Work on internal Mad Devs projects and engineering tools
Investigate engineering issues identified by Enji Guard
Analyze unfamiliar codebases and understand existing technical constraints
Use AI coding agents to implement improvements
Review, test, refactor, and debug AI-generated code
Improve code quality, reliability, maintainability, documentation, testing, CI/CD, and production readiness
Prepare and merge pull requests
Collaborate with mentors and engineering teams through code reviews, technical discussions, and feedback sessions

The goal is not simply to generate code with AI. It is to improve real software systems and make them more reliable, maintainable, and easier to work with.

Requirements:

6+ months of commercial software development experience or equivalent hands-on experience demonstrated through personal, freelance, academic, or open-source projects
Experience using AI coding agents (such as Codex, Claude Code, Cursor, GitHub Copilot, or similar tools) in at least one substantial software project
Experience with at least one of the following programming languages: Python, JavaScript/TypeScript, or C++
A GitHub or GitLab repository, or another project that demonstrates your work
Working knowledge of Git and pull requests
Basic knowledge of Linux, Docker, and relational databases such as PostgreSQL
Familiarity with software quality practices, including testing, debugging, refactoring, and writing maintainable code
Basic understanding of CI/CD concepts and modern software development workflows
The ability to clearly explain what the AI generated, what decisions you made, and why

Additional experience we value:

Contributions to open-source projects
Hands-on work with CI/CD pipelines
Public project deployments
Bug fixing in unfamiliar codebases
Hackathon, bootcamp, or competitive programming achievements
A product, startup, automation, or tool of your own that delivers measurable value

How to apply

Along with your CV, please send us a brief, evidence-based application demonstrating how you work with AI-assisted development. Share links to relevant GitHub or GitLab projects and briefly explain what you built, which AI tools you used, and how you used them. We’d also like to understand what was generated by AI and what you personally designed, reviewed, fixed, tested, or improved. If available, mention testing, CI/CD, deployments, active users, or other real-world usage.