Ivan Magda
Notes on Swift, AI agents, and developer tools.
Featured
-
Foundation Models, Year Two: From On-Device API to General LLM Runtime
WWDC26 turned Foundation Models into Swift's general LLM client: a rebuilt on-device model, a server model with no API keys, and the first real agent primitives.
-
Five Things I Learned About Making AI Coding Agents Actually Work
Updated:The biggest gains in AI agent performance come from the system around the model, not the model itself.
-
Building a Coding Agent in Swift, Part 0: Bootstrapping the Project
Updated:Setting up a two-target SPM package, strict concurrency, and AsyncHTTPClient — the foundation for a CLI agent that never needs restructuring.
Recent Posts
-
Why my Swift debouncer got 50× faster and what the Swift Runtime has to do with it
A Swift debounce investigation through task reuse, runtime metadata, generic specialization, lock placement, and callback latency.
-
Keeping Project Memory Shared Between Claude Code and Codex
How I moved project memory into shared repository files so Claude Code and Codex can read and maintain the same notes.
-
Add Native Progress UI to Your Telegram AI Agent
Show tool calls and progress in Telegram with native thinking blocks, animated action icons, and a Stop button.
-
An Agent Framework Hiding Inside a Session: Dynamic Profiles in Foundation Models
Dynamic Profiles turn one LanguageModelSession into a declarative agent runtime: per-phase models, history transforms, and two orchestration patterns.