Applied AI engineer with an architecture and construction-operations background, currently completing a Master's in Applied Artificial Intelligence at Tecnológico de Monterrey. I design and ship production GenAI systems: multi-agent LLM pipelines, document-intelligence extractors, RAG-style retrieval, and end-to-end automation that connects models to real business workflows.
I don't stop at notebooks. Every project here is containerized, deployed, and running — FastAPI / Node backends, Supabase / SQL data layers, Claude API orchestration, and n8n automation, shipped on a self-managed Docker VPS. My domain edge is turning messy, real-world operational problems into reliable AI-driven products.
What I bring to a GenAI / AI Automation team: LLM orchestration & prompt engineering, retrieval pipelines, ML modeling, API integration, and the full-stack discipline to take an AI feature from prototype to a deployed, monitored, secured service.
| Area | Technologies |
|---|---|
| GenAI / LLMs | Claude API (Opus/Sonnet 4.x), agent orchestration, prompt engineering, prompt caching, vision-assisted extraction, RAG, embeddings / semantic search |
| Machine Learning | PyTorch, TensorFlow/Keras, Scikit-learn, XGBoost, LSTM, CNNs, ensemble methods, feature engineering, NumPy from-scratch backprop |
| Backend / APIs | FastAPI, Express/Node, Python 3.11/3.12, REST, Server-Sent Events (SSE), webhooks, auth & RBAC |
| Automation | n8n (20+ workflows), Twilio WhatsApp Business, Meta WhatsApp Cloud API, Telegram, scheduled pipelines, CRM/ops integrations |
| Data / Databases | PostgreSQL (Supabase + Row Level Security), SQLite / SQLCipher, MongoDB, SQL, Parquet, vector stores |
| Cloud / DevOps | Docker & Docker Compose, self-managed VPS, Caddy reverse proxy, Prometheus + Grafana, Git/GitHub |
| Frontend / Dashboards | React 18 + TypeScript + Vite + shadcn/ui, Vanilla JS SPA, TailwindCSS, Chart.js, Streamlit |
Problem solved. SMBs lose leads that message outside business hours or go cold without follow-up. Zook runs a production sales agent over WhatsApp Business (Twilio) and web chat — with human hand-off, a CRM, and campaign tooling around it.
My role. Sole architect & engineer — agent design, backend, channels, CRM, deployment.
Key features.
Architecture. FastAPI (Python 3.12, 4 workers) + SQLite/Supabase state; Twilio WhatsApp webhooks; n8n for the web channel; Dockerized behind Caddy on a VPS.
Demonstrates: Production LLM-agent engineering (buffering, multi-worker state, reply shaping), grounded retrieval, channel integrations, end-to-end ownership. Code: private repository — walkthrough available on request.
Problem solved. Assembling a public-works bid is a ~5-day manual marathon of reading hundreds of PDF pages, extracting requirements and pricing, and producing dozens of compliant annexes. LicitaGen targets ≤1 day.
My role. Product owner & full-stack/AI engineer — extraction pipeline, generators, OPUS ERP bridge, security architecture.
Key features.
Status: v1 in daily internal use at an active construction firm (desktop deployment: extraction + generators + OPUS bridge); v2 platform (LLM extractor, price ML, security hardening) specified and in iterative build.
Demonstrates: Real-world document intelligence, multimodal LLM fallback design, RAG/embedding retrieval, ML dataset engineering, security-conscious deployment. Code: private repository — walkthrough available on request.
Problem solved. Businesses need automated WhatsApp conversations plus human oversight, multi-bot management, and secure multi-user access.
My role. Full-stack engineer — React dashboard, Supabase data/security model, WhatsApp webhook, n8n automation.
Key features. AI-agent behavior configuration panel (response parameters, bot enable/disable); role-based auth + admin panel; multi-bot management; real-time conversation inbox with tags/notes/quick-replies; Meta WhatsApp Cloud API webhook; n8n automated responses; 12-table PostgreSQL schema with non-recursive Row Level Security.
Architecture. React 18 + TypeScript + Vite + shadcn/ui frontend; Supabase (PostgreSQL + Auth + RLS) backend; Express webhook server; n8n automation layer; Dockerized on VPS behind Caddy.
Demonstrates: Full-stack AI automation, secure multi-tenant data modeling, third-party API integration, workflow orchestration. Code: private repository — walkthrough available on request.
Problem solved. Converts noisy market data into actionable, confidence-scored directional signals delivered automatically.
My role. ML engineer — data pipeline, modeling, API, monitoring, automation.
Key features. Ensemble of an LSTM price model and an XGBoost direction classifier; MetaTrader 5 live data; technical-indicator feature engineering (RSI, MACD, Bollinger, ATR); authenticated FastAPI; confidence-thresholded signal tiers; n8n → Telegram alerts every 5 min; Prometheus + Grafana monitoring.
Demonstrates: MLOps end-to-end, ensemble modeling, time-series, API-first ML, automation integration.
Problem solved. Construction firms juggle budgets, expenses, estimates, schedules, RFIs, payroll and accounting across disconnected tools. Control de Obra unifies 25+ modules and runs the day-to-day of Supernova Arquitectos, an active construction company (its first tenant), on real budgets, estimates and payroll.
My role. Sole architect & engineer; also the SaaS platform layer (tenants, subscription plans, admin, error monitoring).
Key features. 25+ operational modules (budget, Gantt, expenses, POs, estimates, RFIs, punch list, accounting/CFDI, payroll); custom bcrypt auth; multi-tenant isolation; subscription tiers; centralized client-side error monitoring with fingerprinting; PDF/Excel export.
Architecture. ~20k-line Vanilla JS SPA + TailwindCSS; Supabase (PostgreSQL + RPC) with custom auth/session model; separate SaaS admin console; Dockerized on VPS behind Caddy.
Demonstrates: Large-scale full-stack ownership, multi-tenant SaaS architecture, data modeling, observability — the engineering maturity AI features must live inside.
| Project | Type | AI / ML Component | Main Stack | Deployment | Business Value |
|---|---|---|---|---|---|
| Zook | Production GenAI agent | Claude sales agent — RAG grounding, stage/score tags, follow-ups | FastAPI · Claude · Twilio · Docker | Live (zook.mx) | 24/7 sales agent answering real customers |
| LicitaGen | AI document-intelligence | Claude vision extraction, embeddings/RAG, XGBoost | Node · Python · SQLCipher · OPUS | Internal daily use (desktop) | ~5 days → ≤1 day bid assembly (target) |
| WhisperWind | AI automation / full-stack | LLM/n8n conversational automation | React/TS · Supabase · WhatsApp · n8n | Live (VPS) | Automated WhatsApp w/ human oversight |
| Gold Predictor | ML service + MLOps | LSTM + XGBoost ensemble | Python · PyTorch · FastAPI · n8n | Dockerized (VPS internal) | Automated, monitored ML signals |
| Control de Obra | Full-stack SaaS | Operational data platform (AI-ready) | Vanilla JS · Supabase · Docker | Live (VPS) | Runs an active construction firm daily |
Supporting work: fgconnect-whatsapp-bot (public sanitized extract of a production WhatsApp sales bot for a US ISP retailer — debounced inbound buffer, multimodal ingest, live-editable prompt, funnel tracking, human hand-off), fuel-route-api (public Django REST API — route + cost-optimized fuel stops, single OSRM call, offline geocoding, 16 tests),chatbot-rag-llm-cheatsheets (public RAG chatbot — tabular extraction, FAISS, multilingual embeddings), VecinosPro (HOA SaaS, OpenPay), Inventario MDV (inventory SaaS), 20+ n8n automation workflows, FGCONNECT lead-gen, and graduate deep-learning work (a 620K-parameter CNN trained on CPU reaching 78.7% on CIFAR-10 with 2.5× fewer parameters than an FC baseline; NumPy-only backprop networks).
I build AI systems that run in production, integrate with real business workflows, and create measurable operational value — and I own them end to end, from model to deployment.
2017–2021 — B.Arch, Tecnológico de Monterrey (professional license 12957387).
2019–2023 — Construction & engineering roles (Bioecosfera, Grupo México, Axis Arquitectura, Quinven): site engineering, QA/QC and BIM coordination on projects up to $300M MXN — the operational domain my AI work now automates.
2024–present — Founded Supernova Arquitectos (construction) and built its internal tooling — Control de Obra and LicitaGen — to run real operations, which is where my production engineering discipline comes from.
2025–present — Master's in Applied AI, Tecnológico de Monterrey; designed, shipped and operate Zook's production WhatsApp sales agent.
Availability: I'm looking to join an AI engineering team full-time. Supernova's operation is handled by my business partner (Ing. Daniel Loera), and Zook remains a nights-and-weekends side project. I worked independently because I had real operational problems worth automating — now I want to build with a team, at scale, with code review and production standards beyond my own.