Cloud-native architectures, AI-powered solutions, and scalable platforms — crafted with 15+ years of hands-on engineering expertise.
Luballe is a software engineering firm founded by Luis Ballesteros — a systems engineer with an M.Sc. from Universidad de Los Andes, specializing in visualization, robotics, and high-performance computing.
We bridge the gap between complex infrastructure and elegant user experiences. From GPU-accelerated computer vision pipelines to real-time geospatial dashboards, we deliver production-grade systems that scale.
const luballe = {
founded: "Richmond, TX",
experience: "15+ years",
focus: [
"Cloud Architecture",
"AI & Computer Vision",
"Full-Stack Platforms",
"IoT & Embedded",
"DevOps Automation",
],
industries: [
"Smart Transportation",
"Pharmaceuticals",
"Insurance",
"Government",
"Energy",
],
motto: "Build it right,
scale it fast."
};We transform complex technical challenges into production-ready solutions, from concept to deployment and beyond.
Design and implement scalable cloud-native architectures across Azure, AWS, and GCP. Kubernetes orchestration with AKS, EKS, and GKE. CI/CD pipelines via Azure DevOps, infrastructure as code, and containerized microservices for zero-downtime deployments.
Production-ready machine learning pipelines for real-time object detection, license plate recognition, and traffic analytics. GPU-accelerated inference with CUDA, Numba, and TensorRT. AI-augmented development using Claude sub-agents/MCP servers, GitHub Copilot, and Gemini CLI.
End-to-end web platforms with React/Next.js frontends and Node.js/Python backends. Real-time WebSocket communications, interactive geospatial dashboards with Google Maps API, Mapbox, Leaflet, and Deck.gl. UI wireframing with Figma and responsive design with Tailwind CSS.
Robust data pipelines using Kafka, PostgreSQL replication, and cloud-native data services. Scientific computing with NumPy, Pandas, and SciPy. From extraction to transformation to real-time analytics — built for reliability and scale.
Custom embedded solutions from autonomous drone systems to industrial IoT sensors and environmental monitoring. Real-time control systems with NVIDIA Jetson, Raspberry Pi, Arduino, and STM32 microcontrollers. PCB design and multi-OS deployment (Linux, FreeRTOS, Windows).
Comprehensive system monitoring frameworks detecting hardware failures, anomalies, and performance bottlenecks. Real-time alerting with intelligent event analysis, multi-tier notifications, and automated testing frameworks for CI/CD pipeline validation and performance optimization.
Battle-tested tools and frameworks, selected for production reliability and performance at scale.
A selection of platforms and systems we've architected, built, and deployed in production environments.
A comprehensive real-time traffic monitoring and management platform integrating video analytics, geospatial visualization, and AI-powered alerting. Built for Bogota's Secretary of Transport, serving one of Latin America's largest cities.
Designed and implemented computer vision prototype for automated license plate recognition and traffic analysis using NVIDIA Jetson Xavier NX. GPU-accelerated algorithms with Python, Numba, and CUDA for parallel speed calculations. Optimized pipelines using YOLOv4, Darknet, TensorFlow, and TensorRT for real-time vehicle detection and tracking.
Production vision system extracting real-time vehicle velocities from multi-camera feeds using feature matching and Kalman filtering. Deployed across 15+ GPU-accelerated streams for traffic flow optimization and congestion detection.
Defined software and system requirements for a cloud-native genomic platform at Takeda Pharmaceuticals. Delivered containerized bioinformatics services on AWS with JupyterHub deployment. Integrated AI development tools (Claude, ChatGPT, GitHub Copilot) for optimized code quality and efficiency.
Pilot ITS integrating real-time data from Waze API and traffic cameras. Full-stack with Next.js frontend, Node.js backend, Oracle and PostgreSQL databases. Scalable deployment on Azure AKS with comprehensive CI/CD pipelines. UI wireframes designed in Figma.
Production-ready PostgreSQL replication tool featuring atomic dump-based and incremental sync modes, automatic safety backups, foreign key integrity verification, and 68+ table support. Optimized to 2.1 min for 206 MB.
Full-stack geospatial platform for urban mobility analytics. Next.js frontend with Google Maps API, Radix UI components, and authentication. Python backend with advanced geospatial processing using Shapely and PyProj.
Complete IaC pipeline for Azure infrastructure: AKS clusters, PostgreSQL databases, HPA/KEDA auto-scaling, Azure Monitor integration, Key Vault secrets management, and K6 load testing — all automated through CI/CD.
Built autonomous drone systems from core hardware: flight controllers, IMUs, ESCs, and camera systems. Real-time control with C++ and Lua, mission planning with DroneKit Python, ArduPilot, and STM32-Nucleo. Defined functional requirements and testing protocols for safety-critical operations.
Developed scalable backend services using Node.js and Python with MySQL and PostgreSQL. Built responsive React.js frontend with UI/UX optimization. Deployed on AWS Elastic Beanstalk and Amplify with automated scaling and monitoring.
Comprehensive Linux monitoring framework detecting unclean shutdowns, hardware failures, and anomalies through systemd service integration, journalctl event analysis, and automated 30-minute health check cycles with multi-tier alerting.
A collection of open-source embedded projects: MQ2 gas detector, electrovalve control with RTC scheduling, electric pump automation, and environmental monitoring with Raspberry Pi. Arduino-based hardware with custom PCB integration.
Whether you need a cloud architecture review, an AI-powered prototype, or a full production platform — we're here to help turn your vision into engineered reality.