Contributing to Open Source Projects

Why Contribute?

Recruiters Notice:

• Shows collaboration skills
• Proves you can read and understand others' code
• Demonstrates Git workflow knowledge
• Builds network in ML community

Career Benefits:

• Get code reviewed by experienced engineers
• Learn production code patterns
• Gain references from maintainers
• Stand out from candidates with only personal projects

Finding Good First Issues

ML/AI Projects to Contribute To:

• HuggingFace Transformers - Label: good first issue
• scikit-learn - Label: easy, good first issue
• LangChain - Label: good first issue
• FastAPI - Label: good-first-issue
• Streamlit - Label: feature, good first issue

Where to Find Issues:

1. GitHub search: "label:good-first-issue language:python ml"
2. goodfirstissue.dev
3. up-for-grabs.net
4. Project's CONTRIBUTING.md file

Types of Contributions

Start Small:

1. Documentation (easiest entry)

   • Fix typos
   • Improve examples
   • Add missing docstrings

2. Bug Fixes

   • Reproduce reported bugs
   • Write test cases
   • Fix and submit PR

3. Feature Additions

   • Add small utilities
   • Improve error messages
   • Add type hints

Avoid:

• Major refactors (too risky for first PR)
• Breaking changes
• Large feature additions

Contribution Workflow

# 1. Fork and clone
git clone https://github.com/YOUR_USERNAME/project.git
cd project

# 2. Create branch
git checkout -b fix-typo-in-readme

# 3. Make changes, commit
git add .
git commit -m "docs: fix typo in quickstart guide"

# 4. Push and create PR
git push origin fix-typo-in-readme

Writing a Good PR

PR Title Format:

type: brief description

Examples:
docs: add example for fine-tuning BERT
fix: resolve memory leak in data loader
feat: add support for custom tokenizers

PR Description Template:

## What does this PR do?
Fixes #123 - Adds example for fine-tuning BERT on custom dataset

## Why is this needed?
Current documentation doesn't show how to fine-tune models

## Testing
- [ ] Added unit tests
- [ ] Tested locally with example dataset
- [ ] Documentation builds without errors

Etiquette Rules

Before Contributing:

1. Read CONTRIBUTING.md
2. Check if issue is already assigned
3. Comment: "I'd like to work on this"
4. Wait for maintainer approval

During PR:

• Respond to reviews within 24-48 hours
• Be open to feedback
• Keep discussions professional
• Don't take rejections personally

Red Flags to Avoid:

• Submitting PR without issue discussion
• Ignoring code style guidelines
• Arguing with maintainers
• Spamming multiple PRs

:::