I build AI-native developer products with verifiable demos, inspectable artifacts, and bounded failure behavior.
My work spans local-first developer tools, agent infrastructure, and teaching workflows. Each project links a concrete input to an output you can inspect: a flow graph, review queue, teaching replay, or task receipt.
Explore the portfolio · Try Review Pack · Browse repositories
The DESIGN:OS projects below are public forks and extensions. Product-specific licenses and repository notices remain authoritative. The links identify upstream sources where relevant; the descriptions below are limited to the work visible in each NamAIEngineer repository.
| Product | What you can inspect |
|---|---|
| DESIGN:OS Studio | Native Rust/wgpu interface-design studio, .fig read/write, Agents panel, MCP server, Dev Mode code generation, and headless screenshot/diff tooling. This repository follows the upstream project's source-available license and notices. |
| Code2Flow / Review Pack | Local static HTML/code flow graphs plus Review Pack graph-revision comparison, source evidence, reviewer notes, and JSON export. Case study · Release · Interactive demo. |
| Pedagogy / Agent Teacher | Local Vietnamese fraction-teaching rehearsal with authored coaching, transfer, replay, and Markdown export. Case study. The repository retains its own license and notices. |
Watch the 54-second System Agentic AI introduction, or read its captions, transcript, and reproducible composition. The film uses real sanitized provider and decision-plane readbacks from public commit 7b7168a; the advisory decision did not authorize or submit a task, and connector geometry is labeled illustrative.
The Code2Flow / Review Pack preview remains attached to its own project section above. Its demo uses labeled synthetic CRM fixtures and deterministic focus matching; it does not establish semantic impact.
The Review Pack demo uses a labeled synthetic CRM fixture. Focus matching is deterministic and does not establish semantic impact. Agent Teacher uses authored simulations and does not certify teaching competence. Release evidence describes a specific version, not every environment.
System Agentic AI is a provider-neutral workspace framework for shared operating contracts, provider discovery, bounded task execution, and verification. Its README documents the supported setup path, provider boundaries, and reference-license separation.
git clone https://github.com/NamAIEngineer/system-agentic-AI.git
cd system-agentic-AI
python3 scripts/setup.py --check-only
./scripts/setup.sh
./scripts/doctor.shUse the project's own AGENTS.md, README.md, WORKFLOW.md, and agent-brain/operating-contract.md during initialization. Optional references require an explicit command; a cloned or pinned reference is not execution proof.
- Define the input, expected output, trust boundary, and failure behavior.
- Keep source, licenses, upstream notices, and contribution scope visible.
- Link demonstrations to source and versioned releases.
- State what was checked and what remains uncertain.
Identity: NamAIEngineer is my current GitHub handle. DevAngular333 appears in earlier repositories and release tags; those links are retained for continuity.
Contact: Browse public repositories



