This repository powers the public Zhane Grey AI Engineering Mainframe and the Build Doctor diagnostic app. It combines a polished systems portfolio with deterministic Vercel/Next.js build diagnostics, evidence-backed workflow metrics, searchable project proof, and public-safe links to related AI, Web3, RAG, eval, and visual systems.
The site is intentionally evidence-bound. Public copy and metrics are sourced from public GitHub metadata, verified demo URLs, deterministic tests, local evidence summaries, and browser QA screenshots. Raw local logs, private paths, secrets, prompt contents, and private repo data are not exposed.
| Surface | URL |
|---|---|
| Portfolio Mainframe | https://vercel-build-doctor-agent.vercel.app |
| Build Doctor App | https://vercel-build-doctor-agent.vercel.app/build-doctor |
| Case Study | https://vercel-build-doctor-agent.vercel.app/case-study |
| GitHub Profile | https://github.com/zrt219 |
| Evidence Dashboard | https://zhane-grey-evidence-dashboard.vercel.app |
- A Next.js App Router portfolio mainframe with a dark technical command-center UI.
- A live workflow tracker backed by public-safe aggregate evidence summaries.
- Searchable/filterable project directory data for public GitHub and demo surfaces.
- Build Doctor, a deterministic failed-build diagnosis workflow for Vercel and Next.js logs.
- API routes for diagnosis, report generation, optional LLM enrichment, eval fixtures, and integration health.
- Unit, E2E, audit, and browser QA coverage.
- Public-safe evidence summary files under
evidence/public/.
- Hero and primary actions
- Verified stats ribbon
- Live Workflow Events Tracker
- Featured proof systems
- Guided proof path
- Ralphplan workflow map
- Searchable project directory
- AI engineering capability map
- Evidence and QA ledger
- Contact
Latest public-safe metric snapshot: 2026-07-05
| Metric | Value | Source Label |
|---|---|---|
| Workflow events | 1,331,820 | Workflow Events Tracker |
| Daily delta | -2 | Workflow Events Tracker |
| Codex sessions | 1,010 | Workflow Events Tracker |
| Unique thread ids | 1,010 | Workflow Events Tracker |
| JSONL files | 1,010 | Workflow Events Tracker |
| Corpus size | 14.6 GB | Workflow Events Tracker |
| Source-code lines | 676,615 | Workflow Events Tracker |
| Sessions updated today | 3 | Workflow Events Tracker |
| Public GitHub repos scanned | 34 | GitHub Source Memory |
| Foundry projects | 18 | Daily Evidence Report |
| Solidity files | 326 | Daily Evidence Report |
| AI/RAG/agent files | 13,550 | Daily Evidence Report |
| Generated exports | 8 | Daily Evidence Report |
Public-safe evidence files:
evidence/public/live-workflow-events-tracker.mdevidence/public/daily-evidence-report-2026-07-05.mdevidence/public/session-index-summary.mdevidence/public/github-profile-source-memory.md
| Project | Focus | Links |
|---|---|---|
| Zhane Grey Evidence Dashboard | Evidence organization, claim grounding, public proof surface | Demo / Repo |
| Vercel Build Doctor Agent | Deterministic build-log diagnosis, redaction, report export | Demo / Repo |
| AI Resume Tailor Service | Evidence-aware resume and packet generation workflow | Demo / Repo |
| Resume Evidence RAG Auditor | Resume claim verification and RAG/eval discipline | Demo / Repo |
| AI Gateway Failover Playground | Provider routing, fallback behavior, request traces | Demo / Repo |
| Enterprise Agent Workflow Studio | Approval-gated agent workflows and risk scoring | Demo / Repo |
| DatumX | AI data provenance, validation, and on-chain lineage | Demo / Repo |
Build Doctor turns failed Next.js and Vercel build logs into a traceable diagnostic report.
Workflow:
- Paste or load a build log.
- Redact secrets.
- Classify the failure with deterministic rules.
- Extract evidence lines and affected subsystem signals.
- Generate a local diagnostic trace, safe patch draft, suggested solutions, and markdown incident report.
- Optionally request a sanitized DeepSeek review through OpenRouter. The deterministic diagnosis remains the source of truth.
Primary implementation paths:
src/lib/redact-secrets.tssrc/lib/log-parser.tssrc/lib/failure-taxonomy.tssrc/lib/patch-recipes.tssrc/lib/build-doctor/index.tssrc/lib/build-doctor/llm/openrouter-deepseek.tssrc/app/api/diagnose/route.tssrc/app/api/enrich/route.tssrc/app/api/report/route.ts
Portfolio Mainframe
-> typed project data
-> public-safe evidence summaries
-> workflow metric snapshot
-> responsive proof UI
-> browser QA screenshots
Build Doctor
-> redact secrets
-> classify failure
-> extract evidence
-> map remediation
-> draft report
-> optional sanitized LLM review
- No raw local Codex logs are rendered in the public UI.
- No raw JSONL contents are rendered in the public UI.
- No private absolute paths are rendered in the public UI.
- Secret-like values are redacted before diagnosis output, optional provider review, and report export.
- Optional LLM enrichment fails closed when disabled, missing credentials, rate-limited, or invalid.
- Integration health routes return public-safe readiness/fallback state only.
- Deployment upload ignores local
.env,.codex,.omx,.qa, prompt, and private evidence artifacts.
Latest local verification after the portfolio mainframe, tracker, QA, and wording passes:
| Command | Result |
|---|---|
npm run typecheck | PASS |
npm test | PASS, 32 tests |
npm run build | PASS |
npm run test:e2e | PASS, 15 tests |
npm run audit:security | PASS |
npm run audit:45k | PASS |
npm run audit:report | PASS |
npm audit --audit-level=moderate | PASS, 0 vulnerabilities |
git diff --check | PASS, line-ending warnings only |
Browser QA captured:
- desktop/tablet/mobile portfolio screenshots
- live workflow tracker screenshots
- public copy screenshots
- command palette and evidence-ledger checks
- rendered-page private-path and banned-copy scans
The local browser QA harnesses and screenshots are intentionally ignored from Git and deployment upload.
npm install
npm run devOpen:
http://localhost:3000/http://localhost:3000/build-doctor
Useful commands:
npm run typecheck
npm test
npm run build
npm run test:e2e
npm run audit:security
npm run audit:45k
npm run audit:report
npm audit --audit-level=moderateDeterministic Build Doctor mode works without paid model access. Optional LLM enrichment is disabled by default and uses sanitized diagnosis JSON only when explicitly configured.
Use .env.example for variable names and defaults. Do not commit local .env files.
Production target:
Recommended deploy gate:
npm run typechecknpm testnpm run buildnpm run test:e2enpm audit --audit-level=moderatenpx vercel --prod
| Path | Purpose |
|---|---|
src/app/ | App Router pages and API routes |
src/components/portfolio/ | Portfolio mainframe UI components |
src/data/ | Typed public portfolio data and evidence-backed stats |
src/lib/build-doctor/ | Build Doctor diagnosis and optional provider orchestration |
src/test/ | Vitest coverage |
e2e/ | Playwright browser coverage |
evidence/public/ | Public-safe evidence summaries |
ai-engineering/ | Verified engineering log |
- Build Doctor diagnoses pasted/build-log evidence; it does not inspect a live repository automatically.
- Optional provider review may be unavailable or rate-limited.
- Suggested solutions are deterministic guidance and should be reviewed before code changes.
- Public workflow metrics are dated snapshots, not a real-time stream.
- Local-only evidence is labeled and not overstated as a live demo.
Built and hardened a public AI engineering portfolio mainframe with evidence-backed workflow metrics, searchable project proof, responsive SVG analytics, deterministic Build Doctor diagnostics, safe optional LLM enrichment, public/private evidence boundaries, and verified TypeScript, unit, E2E, audit, and browser QA coverage.
Core XRPL EVM systems plus related public product and AI repositories from the same portfolio.
| Project | Description | Status |
|---|---|---|
| ZUC Mine Command Center | On-chain uranium mining operations dashboard with real-time reserve tracking, miner registry, and direct contract interaction through a frontend-only control surface. | Live |
| U235 Fuel Cycle | Deterministic XRPL EVM fuel-cycle pipeline that tracks uranium batches from ore to enriched fuel rod with full on-chain traceability. | Live |
| ISR Network | In-situ recovery control system with on-chain asset tracking, lifecycle state transitions, and operator-facing industrial simulation. | Live |
| Dark Matter Farm | XRPL EVM staking protocol with three orbit tiers, lock-period yield mechanics, and event-driven reward emissions. | Live |
| Cohr Lab | Semiconductor laser fabrication lifecycle modeled as an immutable on-chain state machine from crystal growth to final pigtail. | Live |
| ForgeX | Foundry-powered XRPL EVM deployment console that combines a natural-language UI, Node CLI orchestration, and realtime shader-based visuals. | Live |
| DatumX | Verification protocol for AI-transformed industrial data with deterministic lineage, validator review, and XRPL EVM finalization. | Live |
| Ethex Lottery Game | Foundry plus Next.js betting workflow that modernizes the EthexLoto lifecycle for XRPL EVM reviewer-facing execution. | Public Repo |
| 3DMoonX | Cinematic lunar industrial-base experience that combines Blender source assets with a React Three Fiber web runtime. | Live |
| Unknown002 | Browser-based 3D engineering viewer for a nuclear-electric propulsion spacecraft concept with staged prompt-pack support. | Public Repo |
| AI Engineering Evidence Engine | Interactive evidence dashboard that turns local engineering proof into a reviewer-facing systems narrative. | Live |
| Build Doctor | Codex-style build diagnosis harness for failed Next.js and Vercel builds with deterministic failure analysis. | Live |
| AI Gateway Failover Playground | Public-facing sandbox for request routing, provider fallback, and resilient AI gateway behavior. | Live |
| Enterprise Agent Workflow Studio | Public-facing studio for approval-gated enterprise agent workflows, risk scoring, and audit-oriented design. | Live |
| Resume Evidence RAG Auditor | Public-facing proof surface for claim verification, evidence retrieval, and grounded resume bullet generation. | Live |
| AI Resume Tailor Service | Static Vercel-ready application for evidence-backed resume, cover-letter, and job-packet tailoring. | Live |
| Fuji | Cinematic Next.js Fuji gallery atlas for portfolio storytelling and visual system design. | Live |
| AI Agents for Beginners | Lesson repository for getting started building AI agents. | Public Repo |
| Agentic RAG Memory Digital Twin Edge System | Public-facing landing page for an agentic RAG, memory, and digital-twin edge-system portfolio project. | Live |