From code to prompts: A developers shift to AI-powered workflows

Tips for developers getting started with AI tools and how to work around LLM limitations.

David, a full stack engineer and Chief AI Officer at Nex AI Advisors, sat down with the All Things Open team to share how AI has fundamentally changed the way he approaches software development. Rather than writing repetitive code and debugging, he now spends time refining prompts and clearly communicating what he wants AI to do. This shift from coding to prompt engineering has made his workflow more efficient and strategic.

Read more What is prompt engineering?

His connection to open source began during college in Linux labs and has since evolved alongside AI. David sees open source as essential in today’s AI landscape, particularly because of its transparency and security benefits. With open source, developers can inspect code, build trust, and collaborate more freely, which is critical in an era where AI is raising safety and ethical concerns.

David highlights that AI can strengthen open source communities by making onboarding new contributors easier. AI tools can explain codebases, assist with code reviews, and create beginner-friendly tasks. These features lower barriers for new contributors and support collaborative growth, aligning perfectly with the open source ethos.

For developers new to AI, David recommends starting with clear goals and experimenting on personal projects. Begin with small tasks, like a basic “hello world” project, and scale from there. He also advises working around LLM limitations by supplementing prompts with the latest documentation to ensure up-to-date results. Empathy, context, and analogies, he notes, help make AI accessible to both technical and non-technical audiences.

Read more: Why AI won’t replace developers

Key takeaways

• AI shifts development from coding to communication – Prompt engineering is becoming a vital skill for developers as AI takes over repetitive tasks.
  
• Open source and AI are mutually reinforcing – Transparency, collaboration, and ease of onboarding are enhanced when AI is integrated into open source projects.
  
• Start small and use documentation – To overcome outdated model knowledge, developers should include updated documentation in prompts and begin experimenting with AI on personal projects.

Conclusion

David’s perspective shows how AI can be a practical and powerful tool when used thoughtfully. By combining open source principles with AI’s capabilities, developers can create more inclusive, efficient, and forward-thinking projects. His advice? Don’t fear AI. Get hands-on, stay curious, and treat it like any other tool that helps you build better software.

Check out the entire We Love Open Source podcast series.

More from We Love Open Source

• What is prompt engineering?
• Why AI won’t replace developers
• Comparing GitHub Copilot and Codeium
• How I use AI agents to automate my workflow and save hours