I am a Senior Reliability Engineer with 30+ years of experience in heavy industry, now building production AI and IIoT systems at the convergence of Operations Technology (OT) and Information Technology (IT).
My focus is bridging the gap between physical machinery and data-driven insights through:
- 🤖 Applied AI / GenAI: Production RAG over technical manuals, LLM-powered tools, and agentic automation embedded in real operational workflows.
- 🏗️ IIoT Architecture: Prototyping industrial data pipelines from sensor to dashboard.
- ☸️ Cloud-Native Infrastructure: Orchestrating bare-metal Kubernetes (K3s) for high-availability industrial apps.
- 📊 Predictive Analytics: Leveraging a Post-Grad in Data Science to turn vibration and sensor data into reliability KPIs.
My Philosophy: I don't just monitor assets; I build the infrastructure that makes assets talk.
🚀 Flagship Project: SpectraIO — Industrial Vibration Observability Platform
From wireless vibration sensors to real-time FFT diagnostics — in the browser. Status: private beta.
Condition monitoring of rotating equipment generates high-frequency vibration data that most plants can't exploit: vendor portals lock the data in, and generic dashboards choke on spectral payloads. SpectraIO ingests a wireless sensor fleet end-to-end and serves trend and spectrum analysis to the reliability engineer in seconds — self-hosted, vendor-neutral, built on open infrastructure.
- Pipeline: Triaxial BLE sensors → IIoT gateway → MQTT 5.0 (HiveMQ) → typed ingester microservice → InfluxDB 3 (Flight SQL) → Next.js + TypeScript frontend.
- Analytics: Interactive FFT spectrum workbench with bearing fault-frequency markers (BPFO / BPFI / BSF / FTF), harmonic cursors, and per-axis trends — canvas-rendered (Apache ECharts) to handle thousands of spectral bins.
- Product-grade: Threshold alerting, issue lifecycle management, role-based access, white-label theming; zero inbound ports via Cloudflare Tunnel; fleet observability with Prometheus + Grafana.
- 📂 Product showcase & architecture → · Application source is private — live demo available on request.
A production Retrieval-Augmented Generation system that lets a plant maintenance team query 876 technical PDF manuals (~40,000 pages) in seconds instead of 20–30 minutes.
- Hybrid retrieval: pgvector semantic search + PostgreSQL full-text fused via Reciprocal Rank Fusion; GPT-4o Vision describes technical diagrams; every answer cites its source document and page.
- Stack: Python (Flask), React + TypeScript, PostgreSQL + pgvector, OpenAI API — Docker on bare-metal K3s behind Cloudflare Tunnel.
- Repository is private (proprietary to current employer) — live demo available on request.
| Project | What it is |
|---|---|
| access360-5-ways-to-visualize | Five direct-to-HiveMQ ways to visualize a wireless vibration monitoring fleet (MQTTX, Grafana Live, Node-RED FFT/waterfall, and more). |
| mpu9250-imu-3d-visualization | Real-time 3D IMU orientation: ESP32-S3 + MPU-9250 streamed over WebSocket, rendered with Three.js, Madgwick sensor fusion. |
| firefly-imu-dashboard | 3D IMU orientation dashboard over BLE/Web Bluetooth on an Arduino Nano 33 BLE Sense with 9-DOF fusion. |
| grafana-proxmox-influxdb3-sql | Single-pane Proxmox VE dashboard for Grafana, built natively for InfluxDB 3 SQL/Flight — node gauges + all VMs/LXC with activity rankings. |
| dotfiles | My portable terminal environment (zsh + Oh My Posh + fzf) — one-command setup for Fedora and Debian/Ubuntu. |
| Category | Tools & Technologies | Proficiency |
|---|---|---|
| Infrastructure | Proxmox VE, Kubernetes (K3s), Docker, Rancher | Expert |
| IIoT Stack | HiveMQ (MQTT), Node-RED, LoRaWAN, ESP32 | Expert |
| AI / ML | RAG (pgvector + RRF), LLMs / OpenAI API, embeddings, MCP, scikit-learn | Advanced |
| Data & Integration | Python, InfluxDB, PostgreSQL, CloudNativePG, n8n | Advanced |
| Visualization | Grafana, Streamlit, PowerApps | Advanced |
| Reliability | Vibration Analysis, Laser Alignment, Predictive Maintenance | Expert |
- LinkedIn: joseangelcedeno
- Web: geekendzone.com