AI Assistants for C++ Developers
summary
Demonstrates GitHub co-pilot chat and inline completions actually editing a real project, swapping C++ REST SDK for Poco by editing vcpkg/CMake and building, and debugging until a running HTTP server responding "Hello World" proves the migration works. Shows prompt engineering tips, token/context limits, and flaky retrieval of surrounding code (missing headers, incomplete context). Notes LLM growth (dozens of recent models, hundreds of billions of parameters) and stresses developer responsibility for correctness.
AI-written summary · about
chapters (5)
- 0:00Introduction to AI
- 1:01Understanding LLMs
- 7:09Prompt engineering
- 12:57Practical C++ demonstration
- 41:33Responsible AI principles
related talks
31mEnhancing C++ development with Copilot Chat
28mMaking unfamiliar C++ code familiar with GitHub Copilot
05mco_pilot
39mAI Assistance Beyond Code: What Do We Need to Make it Work?
34mNavigating AI assistant development
46mPragmatic techniques to get the most out of GitHub Copilot
15mVibe Coding at Scale: Customizing AI Assistants for Enterprise Environments
25mImproving developer happiness with AI