Field Notes & Research
Observations, deep dives, and working notes from active research.
12 Papers on Agent Runtimes: What Worked, What Didn't
Michael Hofweller•Apr 15, 2026•18 min read•PublishedWe surveyed 12 recent papers on agent runtimes while designing a TypeScript kernel. The strongest finding: every part of an agent you treat as static — tool catalogs, context, memory, identity — costs measurable accuracy or measurable tokens, often both.
Subagent Driven Development: What the research actually shows
Michael Hofweller•Apr 2, 2026•12 min read•PublishedAfter months of subagent-driven development, I noticed error rates climbing and code coherence dropping — even on simple projects. I went looking for why. The research confirmed what I was seeing and changed how I think about delegation in our agent architecture.
Per-Project Cost Tracking for an Agent-Native Lab
Michael Hofweller•Mar 26, 2026•9 min read•PublishedWe built a per-project cost counter so we can give collaborators real estimates and calibrate intelligence allocations across models to optimize ROI by goal importance.
The Seven Unknowns: What AI Still Cannot Solve in 2026
Michael Hofweller•Mar 22, 2026•18 min read•PublishedAI capabilities are advancing faster than our ability to understand, verify, control, or govern them. These are the large, unresolved problem spaces defining the field in 2026.
Research Agenda: Five Questions Driving the Facility
Michael Hofweller•Mar 17, 2026•5 min read•PublishedWhat does the operating system for an agent-first organization look like? The major questions driving the facility.
The Incongruency Problem: Why AI Is Failing Enterprise
Michael Hofweller•Mar 7, 2026•10 min read•PublishedWe're bolting AI onto a world designed for humans.
From Aesthetic to Algorithm: Building the LORF Design System as an Agent Skill
Michael Hofweller•Feb 14, 2026•7 min read•PublishedHow we turned a visual identity into machine-readable instructions that AI agents can execute.