How to get started with agentic AI using open source tools

Agents vs assistants explained: Why the difference matters for developers.

Lee Faus, Global Field CTO at GitLab, sat down with the All Things Open team to share why agentic AI is exciting, how data driven engineering is shaping development, and what developers should focus on in this fast-changing AI landscape.

Lee’s journey as both a teacher and a technology leader shapes how he thinks about software and AI today. At GitLab, he works with executives to connect developer experience with business outcomes. He shared how data driven engineering, including practices like FinOps, helps teams understand the cost and impact of every change. Lee explained how GitLab even ties merge request activity back to financial accountability, ensuring development work drives real value for users, customers, and the community.

When the conversation shifted to agentic AI, Lee broke down the difference between assistants and agents. Assistants follow specific instructions, while agents operate with more autonomy, similar to a sports agent who negotiates on behalf of a player. He highlighted emerging patterns like swarms of agents, which can combine the strengths of different language models, and pointed developers to tools like LangChain, LangGraph, and LangFlow as starting points for experimenting with open source frameworks in this space.

Read more: Want to get into AI? Start with this.

Drawing from his years of teaching, Lee emphasized that AI is unlikely to replace developers anytime soon. Instead, he sees the role of the developer evolving. Syntax will become less important as agentic AI grows more capable, but critical thinking, collaboration, and understanding architectural patterns will matter more than ever. He encouraged new contributors to jump into open source by writing documentation, opening issues, or tackling “good first issues” on GitHub or GitLab, noting that every contribution, no matter how small, creates an impact.

Key takeaways

• Data driven engineering connects software development to business value, helping organizations manage change more effectively.
• Agentic AI is about more than chat, it’s about autonomous systems that can act on behalf of users, with open source tools offering great ways to experiment.
• Developers should prioritize problem solving, teamwork, and architecture skills, since AI will handle syntax but not human judgment.

Conclusion

Lee reminded the ATO community that the most important practice is to always keep learning. Whether through exploring new frameworks, contributing to open source, or developing critical problem solving skills, the developers who stay curious will be the ones who thrive as AI continues to reshape the field.

More from We Love Open Source

• Want to get into AI? Start with this.
• Getting started with AI on a budget
• The secret skill every developer needs to succeed with AI today
• How I use AI agents to automate my workflow and save hours
• Skip the crowded job hunt: Find your tribe instead