Series
AI Coding Systems
- 01 My Vibe Coding Journey - Developing an AI-Powered English Learning Application
- 02 Building an English Dictionary with an LLM
- 03 ReAct Agents: Building From Scratch - Native Function Calling vs. Custom TAO Parsing
- 04 Automated Prompt Optimization in DSPy: Mechanisms, Algorithms, and Observability
- 05 One Month of Vibe Coding with GitHub Copilot
- 06 GitHub Copilot CLI System Prompt The complete system prompt for GitHub Copilot CLI - a terminal assistant built by GitHub to help with software engineering tasks
- 07 GitHub Copilot Model Token Speed Benchmark: Decomposing Latency, Verbosity, and Throughput Across 28 Hosted Models An empirical analysis of 280 sequential benchmark runs across 28 GitHub Copilot-hosted models and 10 scenarios — separating the 8–14s first-token prefill floor from true marginal token streaming speed, verbosity bias, reasoning effort overhead, and polyglot coding performance.