Perfil GitHub de Simon Gonzalez De Cruz
simongonzalezdc/simongonzalezdc
Construtor de MCP servers e sistemas de IA com 11 anos em ops empresarial
Ver no GitHubPerfil profissional de um desenvolvedor especializado em MCP servers, AI systems e agent tools. Mostra portfólio de trabalhos em Python e TypeScript, com foco em automação empresarial e ferramentas de IA producentes. Fundador da Kyanite Labs e consultor em PuenteWorks.
Instalação
npx convez-ai simongonzalezdc/simongonzalezdc1 comando no terminal: detecta o formato do repositório (marketplace ou plugin avulso) e instala via Claude Code. Requer Node 18+ e o Claude Code instalado.
Métricas
Casos de uso
- Explorar portfólio de MCP servers e agent tools
- Buscar desenvolvedor especialista em automação de IA
- Estudar padrões de código em Python/TypeScript para IA
Veredito

Perfil autêntico de engenheiro full-stack focado em MCP servers produtivos. Valor real está nos repositórios vinculados (Kyanite Labs, PuenteWorks), não neste README.
README
A tradução para o português deste README ainda está na fila.
I build MCP servers, agent tools, and domain software for workflows that need proof — not just prompts. Currently at Kyanite Labs (open-source AI tools) and PuenteWorks (consulting). 11 years in enterprise ops before I started shipping code full-time, which is why I care about the parts demos usually skip: install paths, docs, tests, edge cases, and whether a human can understand what the system is doing.
GitHub Stats
Tech Stack
Featured Projects
| Project | What it does |
|---|---|
| mcp-video | Video editing MCP server — FFmpeg, Hyperframes, repurposing tools. Lets Claude Code, Cursor, and other agents inspect and cut video locally. |
| DialectOS | Spanish dialect localization MCP server. Translate and QA across regional variants, register, and product copy. Published as @espanol/mcp. |
| Epoch | Time/cost/risk estimation MCP server for AI agents. PERT, Monte Carlo, sprint forecasting, token-to-time mapping. |
| OpenGlaze | Ceramic glaze calculator, UMF analyzer, CTE estimator, and recipe manager for potters. |
| checkyourself | Production-readiness auditor for AI-built apps: evidence-based 0–100 score, guided fixes, learning plan. |
| tastecheck | 15 frontend craft skills for AI coding agents. Audits typography, color, states, a11y, and anti-slop. |
| dev-learning-archaeologist | Git-history learning diagnostic. Turns repo history into evidence-backed learning plans. |
Try Epoch
claude mcp add epoch -- npx @kyanitelabs/epoch
Use it on real tasks. Share anonymized estimated-vs-actual feedback so the calibration gets sharper. That's the whole loop.
Background
- Kyanite Labs — open-source AI tools, MCP servers, domain software
- PuenteWorks — AI consulting and workflow automation for small businesses
- Capital Group — talent development, Workday Learning, M365 Copilot adoption
- Southern California Edison — project/program analysis, SAP SuccessFactors
- PIMCO / American Red Cross — training operations, enablement, bilingual instruction
Where to Start
- Clearest proof? mcp-video, Epoch, DialectOS
- Company surface? Kyanite Labs org or kyanitelabs.tech
- How I work? Read the docs, tests, release notes, and launch checklists in any of the pinned repos. That's where the quality signal lives.