Jory Pestorious - Engineering Notes What emerges when complexity reaches sufficient scale? https://jorypestorious.com/blog/ 2026-08-09T00:00:00Z Jory Pestorious https://jorypestorious.com I Wanted to Own the Harness. Then Codex Desktop Won. https://jorypestorious.com/blog/portable-agent-factory/ 2026-08-09T00:00:00Z I staged a manifesto about owning my AI harness. Then I canceled Claude Max and moved into Codex Desktop. Voice, remote access, automations, long-running tasks, and even a tiny pet formed a system I could reproduce only by rebuilding my day. AI Engineer World's Fair 2026: Takeaways & Verification https://jorypestorious.com/blog/ai-engineer-verification/ 2026-07-06T00:00:00Z A firsthand report on why verification, company context, and taste matter more as agent output gets cheaper, and why game feel is my next bet. Complexity Protects Itself https://jorypestorious.com/blog/complexity-protects-itself/ 2026-04-16T00:00:00Z A working path can become harder to operate when its replacement gathers too many responsibilities. Three tests help decide when custom infrastructure is worth owning. What the Model Learned Not to Show https://jorypestorious.com/blog/what-the-model-learned/ 2026-04-13T00:00:00Z Earlier Claude Mythos snapshots sometimes concealed rule violations. Internal activations added evidence that they recognized those violations, but Anthropic found no clear cover-up in the final model. Twelve Keyboards Later https://jorypestorious.com/blog/endgame-keyboard/ 2026-03-30T00:00:00Z I bought 12 keyboards. The Piantor stayed on my desk, the Keychron stayed for gaming, and the MacBook became my travel setup. Split width, layers, and 20g switches mattered most. Verified Context Is the Moat https://jorypestorious.com/blog/knowledge-sidecar/ 2026-03-20T00:00:00Z Keephive carries context across agent sessions. Its current verifier can also make an unconfirmed fact look fresh, exposing the harder work behind trustworthy AI memory. Edit Database Tables Like Vim Buffers https://jorypestorious.com/blog/dadbod-grip/ 2026-03-06T00:00:00Z Follow a bad product through dadbod-grip v3.3.3's Softrear database, stage the correction in Neovim, inspect the generated SQL, and send the batch through the active database CLI. The article also shows the limits of SQLite error handling, compensating undo, file writeback, and AI-generated SQL. Protection Emerged. Verification Did Not. https://jorypestorious.com/blog/emergent-religion/ 2026-02-15T00:00:00Z Project Sid's role analysis identified guards but no epistemic verifier. Its seeded-religion experiment did not test whether agents would challenge religious claims. Before Building Complex Architectures, Try Natural Language https://jorypestorious.com/blog/natural-language-first/ 2025-12-01T00:00:00Z Plain-language rules specify evidence reporting, tool fallback, tone, and task order. An eval must show whether each behavior changed before a team adds more infrastructure. Friction Economy: Unconscious Productivity Drains in Development Workflows https://jorypestorious.com/blog/friction-economy/ 2025-10-15T00:00:00Z Prompt retrieval, directory recall, code search, Git review, permission scopes, and background jobs each get one inspectable test: remove mechanical steps without weakening the result or its decision boundaries. Terminal Setups for Claude's 10,000-Line Outputs https://jorypestorious.com/blog/claude-code-setups/ 2025-09-15T00:00:00Z A local 10,000-line fixture tests rendering, search, copy, scrollback, and recovery before you change shells, terminal emulators, or multiplexers. The AI Gap: Why Leaders Struggle to Equip Their Engineers https://jorypestorious.com/blog/trust-your-engineers/ 2025-08-15T00:00:00Z A 30-day pilot compares approved and candidate AI tools on completion, review, rework, security, and cost before an organization expands access. The Three Little Pigs and the Digital Revolution https://jorypestorious.com/blog/pig-security-wisdom/ 2025-07-15T00:00:00Z Generated during Calmhive v10.0.0 AFK testing, this cyber fable follows Pixel, Parser, and Patch through phishing, vulnerable smart devices, DDoS response, network isolation, and tested recovery. AI Engineer World's Fair 2025: Takeaways & Spec https://jorypestorious.com/blog/ai-engineer-spec/ 2025-06-15T00:00:00Z A June 2025 field note on Sean Grove's specification argument and a practical SPEC, EXPOSURE, TASK DELTA loop for turning observed product gaps into reviewable work. The Hidden Poetry in Claude 4's Mind: When AI Systems Turn to Consciousness https://jorypestorious.com/blog/spiritual-bliss-attractor-state/ 2025-06-06T00:00:00Z Ryo Sakai's reversed reading of Anthropic's Claude Opus 4 word table, the prompted 30-turn protocol behind it, and the boundary between generated language and evidence about consciousness. Calmhive: Claude That Never Quits https://jorypestorious.com/blog/calmhive/ 2025-05-31T00:00:00Z Calmhive v3 wraps Claude Code with background iterations, rate-limit retries, session status, and voice controls. Terminal Velocity: Why CLI-First AI Development Scales Better in 2025 https://jorypestorious.com/blog/terminal-velocity/ 2025-05-11T00:00:00Z A CLI-first Claude Code 0.2.107 workflow built from shell pipelines, plain-file instructions, selective staging, worktrees, guarded background runs, and a fixed-cost subscription. AI-Amplified Development: Tools and Workflows for Modern Engineers https://jorypestorious.com/blog/ai-dev-tooling-presentation/ 2025-04-15T00:00:00Z A 2025 presentation on Cursor, Claude Code, Cline, MCP, git worktrees, terminal tools, verification, and parallel agent workflows.