https://imitation-alpha.github.io/blog/ Notes on agentic systems, ML infrastructure, and tooling. en Wed, 10 Jun 2026 00:00:00 +0000 https://imitation-alpha.github.io/blog/loop-engineering.html https://imitation-alpha.github.io/blog/loop-engineering.html Loop engineering is harness engineering on a timer: schedule, state, spawn, and hard oracles. The hard part is choosing signals the loop cannot game. Wed, 10 Jun 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/recursive-self-improving-ai-rubicon.html https://imitation-alpha.github.io/blog/recursive-self-improving-ai-rubicon.html A spectrum map for self-improving AI: pseudo-labels, rewards, AI judges, generated tasks, harness improvement, and why we are still near the Rubicon. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-self-correction-decoding-workflow-reasoning.html https://imitation-alpha.github.io/blog/ai-self-correction-decoding-workflow-reasoning.html A practical map of AI self-correction: decoding tricks, workflow feedback, and trained reasoning all need real signals, not just reflection. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-agent-harness-engineering.html https://imitation-alpha.github.io/blog/ai-agent-harness-engineering.html Harness engineering for agent builders: tools, permissions, workflows, feedback, memory, and verification loops that make an LLM usable. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-agent-research-work.html https://imitation-alpha.github.io/blog/ai-agent-research-work.html A practical map of how AI agents change research work: cheaper first passes, more verification pressure, and human judgment moving to problem choice and evidence. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-agent-multi-agent-interaction.html https://imitation-alpha.github.io/blog/ai-agent-multi-agent-interaction.html A builder-focused map of multi-agent systems as network topologies: when agents should branch, critique, compete, merge, or stay independent. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-agent-context-engineering.html https://imitation-alpha.github.io/blog/ai-agent-context-engineering.html A practical taxonomy for agent context: prompt-visible workspace, external memory, compression, bounded observations, sub-agents, and on-demand skills. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/ai-agent-systems-hung-yi-lee-summary.html https://imitation-alpha.github.io/blog/ai-agent-systems-hung-yi-lee-summary.html A computer-architecture map for AI agents: LLM as processor, context as working set, memory as storage, tools as I/O, feedback as diagnostics, and harness as OS. Tue, 19 May 2026 00:00:00 +0000 Arthur Yau https://imitation-alpha.github.io/blog/orchestrating-coding-agents.html https://imitation-alpha.github.io/blog/orchestrating-coding-agents.html How I orchestrate ~10 parallel CLI coding agents with tmux, Tailscale, Termius on iOS, and a small PWA panel for mobile control. Tue, 28 Apr 2026 00:00:00 +0000 Arthur Yau