How I build with AI
I build the workflows around the model: what it can read, who does the work, how we check it and where I make the decision.
AI-built, human-directedAI Orchestrator
A personal AI workspace connecting scoped specialists, source-linked memory, Python checks and human decisions.
Follow the brief through the system. The return route shows a revision.
-
01 Define the outcome Human direction
- Input
- A request, a real constraint and a desired result.
- Output
- A bounded brief with a checkable finish.
This websiteExpand this portfolio's AI projects while keeping the existing brand and private sources separate.
I decide the objective and what the system may change.
-
02 Bound the context Coordinator
- Input
- The brief and the sources permitted for this task.
- Output
- A small assignment for the right specialist.
This websiteUse the public project copy, approved visual system and existing Astro code.
A relevant source does not grant permission to share it.
-
03 Do the work Scoped specialist
- Input
- The assignment, relevant evidence and allowed tools.
- Output
- A draft, analysis or implementation for checking.
This websiteAI agents produce the project text, Astro components and calculated graph on this page.
A specialist's recommendation is advice, not approval.
-
04 Check the result Code and review
- Input
- The result and the task's acceptance conditions.
- Output
- Evidence of what passed, failed or remains unknown.
This websiteThe production build and automated browser checks pass. Visual review remains a separate check.
A successful command does not prove the whole task worked.
-
05 Decide what ships Human review
- Input
- The checked result, its limitations and proposed action.
- Output
- Accept, revise or hold at the relevant release boundary.
This websiteKeep the update local for my visual review before publishing it.
Checks do not replace creative judgement or release permission.
Conceptual workflow, not a live execution trace. Open a stage to inspect the hand-off.
Under the hood
Separate agents do the work. Source-linked memory carries decisions forward. JEV helps with narrow judgements. I connect these parts and keep their permissions distinct.
Multi-agent setup
Personal working system
A coordinator delegates bounded work to separate specialist profiles, then checks the returned evidence.
Agent memory
Working prototype / expanding coverage
Source-linked recall, age-based relevance and corrections that preserve history.
JEV judgements
On-demand advisory tools
Structured choices, scores and yes/no questions support evaluation and routing. Code and people retain control.
Knowledge maps
Owner-only navigation tool
Follow relationships back to the source instead of treating a graph as the answer.
This website
AI-built, human-directed
A real application of the workflow, from creative brief to code and browser checks.


