Learning AI-Assisted Coding with GitHub Copilot

• Wong Ju Wei

Learning AI-Assisted Coding with GitHub Copilot

Recently, I had the incredible opportunity to participate in a training session on AI-assisted coding through the MyTalent TalentCorp programme, managed by AGMO Academy. The session was led by Iszuddin Ismail, and it completely transformed how I think about coding and development workflows.

About the Programme

The MyTalent programme, under TalentCorp, is designed to upskill Malaysian developers with cutting-edge technologies. AGMO Academy's implementation of this programme focuses on practical, hands-on learning that prepares developers for real-world challenges.

What I Learned

1. GitHub Copilot as a Coding Partner

One of the first eye-opening lessons was understanding that GitHub Copilot isn't just an autocomplete tool—it's a collaborative coding partner. Iszuddin emphasized that the key to getting the most out of Copilot is learning how to communicate effectively with it.

Key takeaways:

  • Write clear, descriptive comments to guide Copilot
  • Break down complex problems into smaller, manageable tasks
  • Review and understand all suggestions before accepting them
  • Use Copilot to explore different approaches to the same problem

2. Effective Prompting Techniques

The session taught me that prompting is an art form. The quality of Copilot's suggestions directly correlates with how well you describe what you need.

Best practices I learned:

  • Be specific about requirements (languages, frameworks, constraints)
  • Provide context in comments and variable names
  • Use examples to guide the AI's understanding
  • Iterate and refine prompts based on initial results

3. Code Review and Validation

Iszuddin stressed the importance of not blindly accepting AI-generated code. We learned systematic approaches to:

  • Verify logic and edge cases
  • Check for security vulnerabilities
  • Ensure code follows best practices
  • Test thoroughly before deployment
  • Understand every line of code before shipping it

4. Boosting Productivity Without Compromising Quality

Perhaps the most valuable lesson was learning to use Copilot to accelerate development while maintaining high code quality. This includes:

Practical applications:

  • Generating boilerplate code quickly
  • Writing comprehensive tests faster
  • Creating documentation more efficiently
  • Exploring new libraries and frameworks
  • Debugging with AI assistance

5. Project-Based Learning

We applied these concepts to real projects during the training, including:

  • Building a static site generator (like the one powering this blog!)
  • Creating REST APIs with proper documentation
  • Implementing design patterns with Copilot's assistance
  • Setting up development workflows

The Impact on My Development Workflow

Since the training, my coding workflow has fundamentally changed:

Before Copilot:

  • Spent significant time searching Stack Overflow
  • Slower implementation of repetitive tasks
  • More time context-switching between documentation and code

After Copilot:

  • Faster prototyping and iteration
  • More time for creative problem-solving
  • Better code documentation habits
  • Increased exploration of alternative solutions

Challenges and Learning Curve

It wasn't all smooth sailing. Some challenges I encountered:

  1. Over-reliance: Initially, I found myself accepting suggestions without fully understanding them
  2. Context limitations: Copilot doesn't always have full project context
  3. Learning to prompt: Developing the skill to write effective prompts took practice
  4. Code review discipline: Maintaining vigilance in reviewing AI-generated code

Iszuddin's guidance helped us navigate these challenges and develop healthy habits around AI-assisted coding.

Gratitude and Moving Forward

I'm incredibly grateful for this opportunity through the MyTalent TalentCorp programme and AGMO Academy. Special thanks to Iszuddin Ismail for his patient instruction and practical insights.

The session didn't just teach me how to use a tool—it taught me a new way of thinking about software development. AI-assisted coding isn't about replacing developers; it's about augmenting our capabilities and freeing us to focus on what matters most: solving problems and building great software.

Key Takeaways for Aspiring AI-Assisted Developers

If you're considering learning to code with AI assistance, here are my recommendations:

  1. Embrace the Learning Curve: Give yourself time to adapt
  2. Stay Critical: Always review and understand AI suggestions
  3. Practice Prompting: Invest time in learning effective communication with AI
  4. Maintain Standards: Don't let AI assistance compromise your code quality
  5. Keep Learning: AI tools evolve rapidly; stay updated

The Future

As I continue my journey as a developer, I'm excited to integrate these skills into my freelance work and personal projects. The combination of human creativity and AI efficiency opens up incredible possibilities.

If you're interested in the MyTalent programme or AGMO Academy's training, I highly recommend exploring their offerings. The investment in learning AI-assisted development is well worth it.

About the Programme:

  • Programme: MyTalent TalentCorp
  • Academy: AGMO Academy
  • Instructor: Iszuddin Ismail
  • Duration: Week 3 - AI & Development Tools

Connect with me if you'd like to discuss AI-assisted coding or share your own experiences with GitHub Copilot!