Putting the "fun" in function calling
summary
Shows how function calling (passing formal signatures) replaces brittle JSON parsing and saves teams from fragile regex and retry hacks, with examples: a Blinkops automation that turns a natural language rule into runnable workflows and agents that call APIs and report agent success / failure back as boolean tool calls. Explains that implementations are still prompt engineering all the way, cites a Berkeley paper where tool use boosts accuracy well above prior models (not yet 95%), and outlines practical eval heuristics.
AI-written summary · about
related talks
01h 01mAI Agents In-Depth – Function Calling, MCP and Tool Use Under the Hood
26mContext Engineering: Practical Techniques for Improving Agent Quality Today
40mBuilding Effective AI Agents Simple, Composable Patterns
16mTurning Fails into Features: Zapier’s Hard-Won Eval Lessons
28mWhat You See Is What It Does: A Structural Pattern for Legible Software
48mStop Fighting Your AI Engineering Prompts That Actually Work
25mVirtual Tool Calling and the Future of Portable AI Toolchains
34mProgramming Languages for AI Programing Agents