A Practical Introduction to Agentic Coding

AI Agents have become increasingly good at generating code. Developers who know how to use agentic tools as they program can increase their productivity significantly. In this session, Marlene Mhangami shares how she gets the most out of agents in her development workflow using GitHub Copilot SDK and GitHub CLI. She'll walk through how she uses MCP (Model Context Protocol), Agent Skills and Instructions to create semi-autonomous agents that can complete multi-step tasks end-to-end. We'll walk through practical, copy-pasteable Python code that leverages MCP (Model Context Protocol) servers and Agent Skills to:

•  Work with the M365 suite to send Outlook emails or get information from Onedrive
•  Validate code changes with screenshots from the Playwright MCP server
•  Generate assets like architecture diagrams or brand assets.

This talk is a hands-on case study of agentic coding in action — showing the core mental model (planning + tool use + iteration), effective patterns with today's SDKs, common pitfalls (like MCP setup gotchas), and how these techniques scale to production dev workflows.

Marlene Mhangami

Marlene Mhangami is a software engineer and developer advocate at Voltron Data, where she works on sharing and improving software infrastructure for big data. She’s an advocate for using science and technology for social good. Earlier in her career, she served as the Director and Vice Chair of the Python Software Foundation (PSF), where she helped organize PyCon Africa, the first regional/Pan-African gathering of the Python community. Marlene was recently named Vice Chair of the ACM Practitioner Board, which is responsible for developing programs that support the professional needs of ACM members as well as promoting computing as a profession. She’s based in Harare, Zimbabwe.