InnovAIte is a student-led Capstone Company at Deakin University, dedicated to trialing and validating the AI tools, methods, and frameworks that will power SPARK—Australia's first AI-native startup engine, launching in 2026.

Our mission is to understand and demonstrate how AI can dramatically compress startup development cycles from months to days. By doing so, we aim to make entrepreneurship radically more accessible to everyone, regardless of their technical background.

We operate through two foundational programs designed to explore the full spectrum of AI's potential in an entrepreneurial context.

This program focuses on equipping non-technical users with the practical skills to leverage AI effectively. Through research and workshops, we create educational resources that demystify AI for academic and professional use.

• Key Focus Areas: Generative AI for academic writing, AI-enhanced teaching tools, administrative automation, and prompt engineering fundamentals.
• Key Outputs: We have developed functional prototypes like an AI-powered Thesis Generator (using Cohere), a personalized Research Assistant web app, and a tool for managing AI attribution.

This lab is a hands-on incubator for rapidly developing and testing AI-powered applications. We build Minimum Viable Products (MVPs) to prove how AI can accelerate the journey from idea to a functional prototype.

• Key Outputs: We have successfully built and deployed several platforms, including:
  • NoCodeJam.AI: A live platform for hosting AI-assisted no-code hackathons.
  • InnovAIte Website: The central hub for our company, built with a Supabase backend.
  • Socratic Crumbs: A multimodal chatbot with voice and vision capabilities for enhanced learning.
  • Skill Quest Sprint: A platform for mastering new skills through 30-day guided sprints.

Building on our foundation, we are currently validating a new wave of ambitious projects that span education, accessibility, and impact ventures.

1. Artificial Assessment Intelligence for Educators (AAIE): Developing a custom-trained LLM to assist educators with AI-aware assessment design and feedback.
2. AI Assisted Navigation Device: Prototyping an assistive device to improve mobility and accessibility for people with disabilities.
3. Venture Pipeline Management System: Designing a system to help track and support the growth of inclusive businesses in Southeast Asia.
4. AI Organisation Design: Exploring how AI can transform organizational workflows, with the goal of creating a next-generation AI-powered project planner.

To ensure our collaboration is smooth and our codebase remains clean and stable, please adhere to the following guidelines.

The main branch is our source of truth and is always production-ready.

• Direct pushes to main are not allowed.
• All contributions must be submitted via pull requests (PRs).
• PRs merged into main must have at least one approved review.

All changes—whether code, documentation, or assets—must go through a Pull Request. Every PR should:

• Be linked to an issue or task in our project tracker (where applicable).
• Have a clear, descriptive title and a summary of the changes.
• Be assigned to at least one reviewer for feedback.

Code reviews are critical for maintaining quality.

• Every PR must be reviewed and approved by at least one team member before merging.
• Reviewers should check for:
  • Clarity & Maintainability: Is the code easy to understand and build upon?
  • Alignment: Does it follow our team's coding standards and conventions?
  • Functionality: Does it solve the intended issue or add the feature correctly?

We use a feature-branch workflow to keep our main branch stable.

• ✅ Always branch off from the latest main branch.
• Branch names should be descriptive and follow this convention:

  feature/[feature-name]
  fix/[issue-number]-[bug-name]
  docs/[document-name]
  release/[version-tag]
  

• Once your work is complete, open a PR to merge your branch back into main.

Follow these steps whenever you want to contribute code or documentation:

🔹 Go to the project repository on GitHub.
🔹 Click the Fork button (top right) to create your own copy under your GitHub account.

git clone https://github.com/<your-username>/<repo-name>.git  
cd <repo-name>

This keeps your fork synced with the main project:

git remote add upstream https://github.com/<org-name>/<repo-name>.git

Always create a new branch for your work:

git checkout -b feature/<short-description>

🔹 Edit files / add new code.
🔹 Test your changes before committing.

git add .  
git commit -m "feat: add [suitable commit message]"

💡 Use clear, descriptive commit messages.

git push origin feature/<branch>

🔹 Go to your fork on GitHub.
🔹 Click Compare & pull request.
🔹 Select the base branch as main in the original repo, and your branch as the compare branch.
🔹 Add a description of what you changed and why.
🔹 Request at least one reviewer.

✅ That’s it! Once your PR is reviewed and approved, it can be merged into the main repository.

To protect our repositories, access is managed strictly.

• Admin rights are reserved for team leads.
• All other team members will have collaborator/contributor access.
• Each team must designate one person to hold admin access for their repository and communicate this to the leadership team.

• 💾 Commit Frequently and Meaningfully: Small, atomic commits help us track changes. Use clear, conventional commit messages (e.g., fix: corrected user XP calculation logic). Avoid vague messages like Updated code.
• 📌 Use Issues for Tracking: Open an issue for any bug, feature request, or significant task. Use labels to categorize them (e.g., bug, enhancement, documentation).
• 📚 Keep Documentation Updated: Update the README.md and other relevant documentation whenever setup steps, dependencies, or workflows change.
• 🧪 Test Locally Before Submitting a PR: Ensure your code runs and passes all basic checks (linting, building, tests) on your local machine before opening a pull request.
• 📬 Be Responsive: Reply to code review comments and issue discussions promptly to keep the development process moving forward.

• 📊 Use Curated and Ethical Datasets: Always check the source, license, and potential biases of any dataset. Document all data sources in a DATA_SOURCES.md file.
• 🧼 Document Data Preprocessing: All data cleaning and preprocessing steps must be documented in reproducible scripts.
• 🔒 Respect Privacy: Never commit sensitive data. This includes student info, API keys, or personally identifiable information. Use .gitignore and environment variables for secrets.
• ⚖️ Bias and Fairness Checks: Regularly test models for unfair outcomes across different user groups.

• 🧱 Baseline First: Always start with a simple baseline model or heuristic. This provides a benchmark to measure the performance of more complex architectures.

Our stack includes core development technologies, major AI platforms, and a suite of innovative AI-native tools that allow for rapid prototyping.

All our project code and ongoing work can be explored in our public-facing repositories.

• Explore our Projects on GitLab
• Watch our Demos on YouTube

We are proud to be supported by a dedicated team of academic staff and industry mentors from Deakin University and beyond.