300+ CLAUDE CODE SKILLS · 113 PUBLISHED UNDER MIT · 13.4B TOKENS SHIPPED

Independent AI engineer

Jackson
Lai.

Built with curiosity

Agents.
Tools.
Systems.

I build the systems behind
intelligent agents.

The model isn't the binding constraint anymore. The harness around it is. I build at that layer: decision-first skills, persistent agents, eval-driven shipping.

DRAG TO EXPLORE · CLICK TO MEET

01 / Selected work

Beyond the interface.
Into the engineering.

The skills, agents, and infrastructure that turn an idea into a working system.

All projects

113
Claude Code skills published (MIT)
300+
Skills authored (2 of 5 packs published)
13.4B
Tokens shipped through agents
~80%
Of own code is agent-written
MIT
License — fork freely
Featured · v0.2.0 live

JacksonSuperAI/ai-skills

A Claude Code plugin marketplace. 113 skills in two packs: ai-engineering (86, covering RAG, evals, agents, MCP, fine-tuning, observability, security, multimodal, voice) and b2b-go-to-market (27). Each skill is a 200-300 line decision document with named tools, prices, regulations.

How I think about AI engineering

The harness layer is undervalued.

"The model isn't the binding constraint anymore.
The harness around it is."
  • Better operators ship more than better models do, given the same model. Karpathy's "skill issue" thesis (March 2026) named the thing — the human in the loop is the bottleneck, not GPT-6.
  • The harness layer compounds. Agent files, persistent memory, locked eval surfaces, git-revert ratchets — these are infrastructure investments that pay off across every model upgrade.
  • Decision-first beats template-first. Most "AI skill libraries" are prompt collections. The ones worth installing tell you when NOT to use a pattern.
  • Editorial bar, not aggregational. Rewrite or replace, never reproduce. Surface area shrinks; signal-to-noise improves.
  • Decade of agents, not year of agents. Build for sustained orchestration, not single-shot demos.
Recent thinking

Long-form posts & analyses.

Cross-posted from LinkedIn. The Karpathy series, the X-Men agent-architecture series, and ad-hoc takes on industry news as it lands.

Connect

Always up for harness-thinking conversations.

Open to collaborations, deeper feedback on the library, or talking about agent architecture, eval-driven shipping, and developer-tool design at AI labs.

An interactive energy study

Meet Jarvis.

A new visual identity for my desk agent: light, circuitry, and a little curiosity. Try a visual state, then explore the systems behind it.

A steady core. Ready for the next idea.

Visual demo only. No microphone, chat connection, or live agent data.