Rubber Duck Thursdays - Exploring GitHub Models
Chris explores GitHub Models as an AI prototyping playground, demonstrates Copilot on the command line for learning Linux commands, and shows off a new leaderboard system for the OctoSnap card game. The session also covers two weeks of GitHub changelog updates including fine-grained personal access tokens going GA, enterprise-owned GitHub Apps, instant semantic code search indexing, and Copilot updates across VS Code, JetBrains, Xcode, and Eclipse.
Returning from a week off, Chris opens by demonstrating the GitHub CLI Copilot extension — using gh copilot suggest and gh copilot explain to translate natural language into shell commands, explain complex terminal commands, and learn Linux tooling interactively. He then showcases the OctoSnap leaderboard backed by Cosmos DB on Azure, and uses Copilot agent mode live to fix a bug where usernames with the @ symbol caused issues, including adding backend validation and tests. The second half focuses on GitHub Models, where Chris walks through the model catalog and playground, comparing models like GPT-4o and DeepSeek V3 for a computer science quiz scenario. He demonstrates system prompts, prompt presets, the prompt editor, quick actions for switching between cheaper and faster models, and how to generate integration code for JavaScript and Python. The changelog roundup covers Copilot custom instructions going GA, code referencing in JetBrains, enterprise-owned GitHub Apps, fine-grained PATs reaching general availability, actions performance metrics, and Copilot for Xcode chat becoming generally available.
Related Content
Rubber Duck Thursdays - Building a scoring system
GitHub
Chris introduces OctoSnap, a GitHub-themed memory card game built with Next.js, and dives deep into designing a scoring model with Copilot covering difficulty multipliers, time bonuses, and penalty mechanics.
Use GitHub Spark to create a podcast timer apps
GitHubDemonstrates GitHub Spark on mobile, building a podcast session timer app entirely through natural language prompts. The video covers iteratively adding timer controls (start, pause, reset), a reverse progress bar, a color-coded time-remaining indicator (green/amber/red), and AI-powered timer creation from a plain-language description—showcasing GitHub Spark's built-in LLM integration. It also demonstrates Spark's managed state storage and the share link for giving collaborators read or write access.
Use GitHub Spark to create a travel log app
GitHubDemonstrates GitHub Spark, an AI-native micro-app platform from GitHub Next that lets anyone build, use, and share personalized apps through natural language—without writing code or managing deployments. The video creates a city travel log app that captures reviews with GPS coordinates, plots destinations on a map, generates AI city suggestions based on existing ratings, and uses the variant generator to explore alternate UI directions from an ambiguous prompt. It also covers the built-in theme editor, state management, and revision history for iterating on app design.