Writing Index
Blog
Writing on software engineering, security, architecture, mobile systems, and the occasional adjacent obsession.
Start Here
24 Scenarios That Prove Your AI Agent Follows Rules
Building a governance demo that proves AI agents can be constrained by explicit policy. 24 scenarios with deterministic outcomes and a drift detection exercise.
Navigate Faster
Explore by Topic
Popular Categories
Latest Entries
Recent Writing
Building Dubnium: An Agentic Orchestrator
Jul 13, 2026 12:00AM
8 min read
1645 words
How Dubnium is evolving from local model serving into a governed agentic control plane with a supervisor gateway, bounded memory, specialist routing, and deterministic execution boundaries.
Building Dubnium: Scheduling Agentic Work
Jul 13, 2026 12:00AM
10 min read
2083 words
How Dubnium uses declarative schedules, systemd timers, workflow dispatch, and a separate agentic control plane without turning the scheduler into a hidden workflow engine.
Building Dubnium: A Super Workstation With Modes
Jun 29, 2026 12:00AM
5 min read
999 words
Dubnium is a NixOS-based personal super server built to stay legible: safe defaults, explicit modes, layered mutability, local AI boundaries, and an optional lab surface.
CareerOps: Building a Governed AI-Assisted Job Application Pipeline
May 18, 2026 11:30AM
11 min read
2155 words
A practical architecture for using AI to discuss roles, create job intake issues, trigger local resume generation, and preserve human accountability without turning job search into opaque automation.
Git Is Becoming an Execution Surface
Apr 3, 2026 12:00AM
6 min read
1084 words
Git metadata now crosses execution boundaries in CI and agentic systems, which turns familiar repository fields into potential code execution inputs.