The Work Ledger: Turning Agent Work Into Verified Transactions
Why I added tasks, runs, worktrees, reviews, and plan-bound execution to Soul CLI, and how they turn AI-assisted development into an auditable local system.
Design Technologist,
AI Infrastructure,
Prototyping.
Why I added tasks, runs, worktrees, reviews, and plan-bound execution to Soul CLI, and how they turn AI-assisted development into an auditable local system.
Why Soul keeps registry coordination separate from git, and why remote continuity should mean projection plus promotion rather than distributed writes.
How I built a repeatable link-to-podcast pipeline that extracts a source, writes a spoken two-host script, generates audio, saves notes, and publishes an RSS episode to podcast apps.
How I built an automated, self-healing pipeline that generates a command-line reference directly from the active parser schema without manual translation.
How I structured my local agent environment using isolated execution surfaces, dedicated background runtimes, and a unified file-based session registry.
How I consolidated dozens of loose shell and Python scripts into a single, cohesive command-line tool to manage my agentic workspace.
The amount of work required to implement agents in enterprises will exceed anything we imagine today. Here is why moving from chat to agency is a massive engineering and organizational lift.
The rise of CLI/TUIs isn't about nostalgia. It's about building software that agents can inspect, tweak, and extend—something cloud-locked SaaS simply can't offer.
If the personal stack is going to leave my Mac mini, it needs to be transport-agnostic, easy to pair, and graceful about fallbacks. Here is the shape that recommendation takes.
Tailscale's data plane is peer-to-peer and end-to-end encrypted. The control plane is not. Here is what that actually means, and what Headscale changes.
Two ways to start a design system: ship components and extract tokens later, or define tokens and let components fall out of them. The choice shapes everything downstream.
A speculative prototype I built to test whether you can describe UI by intent and relationships instead of pixel coordinates.
How to separate 'what I learned about the code' from 'what I learned about working with this person' so the AI gets better across projects.
Why I stopped putting AI config files in every repo and moved project orchestration into a centralized registry instead.
Why I stopped treating AI system prompts as per-project config files and started treating them as portable operating systems.