H
Howardismvol. 03 · quiet corner of the web
Howardism · Vol. 03Masthead · No. 00

A wiki, set in plates.

CuratorHoward TaiIssueVol. 03 · no. 128CadenceBatched, not scheduledBasedTaiwan, 23°NContacthoward@howardism.dev

Working notes on interaction-era AI, kept publicly. Each plate is a topic; each topic carries its brightest notes and the source I'm currently digesting. 128 notes across 26 sources, last filed 28 May 2026. Updated whenever a batch finishes — not on a schedule.

Plate · 01
15
filed under interaction
Activity · 8 wks

Interaction, in order.

Real-time multimodal, full-duplex, and human-AI collaboration.

  1. 01AI-Native Product Org Bottlenecks28 May 2026 · 9′
  2. 02The Future of Agent Interfaces28 May 2026 · 11′
  3. 03Does the Human-Facing Harness (HTML Artifacts) Hit Its Own Bloat Ceiling?28 May 2026 · 6′
  4. 04Human-in-the-Loop Boundaries28 May 2026 · 9′
  5. All 15 notes in interaction →
Plate · 02
39
filed under architecture
Activity · 8 wks

Architecture, in order.

Model internals: encoder-free fusion, training, scaling, evals.

  1. 01Agent Identity and Authentication28 May 2026 · 4′
  2. 02Agent Supply Chain Risk28 May 2026 · 5′
  3. 03Agentic Prompt Injection28 May 2026 · 4′
  4. 04How AI-Native Startups Avoid Speed Becoming Strategic Debt28 May 2026 · 9′
  5. All 39 notes in architecture →
Plate · 03
28
filed under harness
Activity · 8 wks

Harness, in order.

Agent loops, tools, orchestration, and Claude Code patterns.

  1. 01Agent Control Plane Patterns: Tickets, Loops, Specs, and Memory Files28 May 2026 · 9′
  2. 02Where Does Agent Harness Work Remain Durable as Models Improve?28 May 2026 · 9′
  3. 03Agent Context Files25 May 2026 · 1′
  4. 04Agent-Native Infrastructure23 May 2026 · 4′
  5. All 28 notes in harness →
Plate · 04
8
filed under alignment
Activity · 8 wks

Alignment, in order.

RLHF, character, interpretability, and safety case studies.

  1. 01AI-Accelerated Offense28 May 2026 · 4′
  2. 02When Does Verification Quality Determine Whether AI Automation Works?28 May 2026 · 8′
  3. 03Agentic Misalignment (AM)8 May 2026 · 4′
  4. 04Alignment Fine-Tuning (AFT)8 May 2026 · 3′
  5. All 8 notes in alignment →
Plate · 05
38
filed under orgs
Activity · 8 wks

Orgs, in order.

Labs, people & products: Anthropic, OpenAI, TML, Claude Code.

  1. 01AI-Native Moats Under Frontier-Model Improvement28 May 2026 · 8′
  2. 02Opinions on Using AI Tools & the Future of the Software Engineering Role28 May 2026 · 12′
  3. 03OWASP28 May 2026 · 3′
  4. 04Zero Trust for AI Agents28 May 2026 · 8′
  5. All 38 notes in orgs →
Plate · ∞

The desk.

The sources behind the wiki, ranked by how often they're cited.

Talk
6
Article
15
Paper
4
Repo
1