Docker

Experience with Docker

Docker is a core tool in Pablo Diaz's development and deployment workflow, particularly for containerizing Python-based microservices, local development environments, and services that require specific runtime configurations not natively supported by serverless platforms. While Pablo's primary web applications deploy on Vercel's serverless infrastructure (which abstracts away containerization), Docker plays a critical role in the supporting services that power the Blossend ecosystem. On OpenMyPro, the AI/ML matching algorithm runs as a containerized Python FastAPI service with specific dependencies (scikit-learn, pandas, numpy) pinned to exact versions — Docker ensures this service runs identically in development, staging, and production regardless of the host machine's configuration. The Dockerfile uses multi-stage builds to separate build-time dependencies from the runtime image, reducing the final container size from ~2GB to ~450MB and improving cold start times. For data engineering pipelines, Docker containers package web scraping, NLP processing, and ETL workflows into reproducible units that can be scheduled and orchestrated independently. Each pipeline stage (crawl, parse, normalize, index) runs in its own container with dedicated resource limits, making it easy to scale individual stages based on their computational requirements. Pablo's Docker expertise includes writing optimized Dockerfiles with multi-stage builds, layer caching strategies for faster rebuilds, docker-compose for multi-service local development environments that replicate the production architecture, health checks for container orchestration, and security best practices including non-root users, minimal base images (alpine/distroless), and vulnerability scanning. His Amazon AWS background provided extensive Docker experience through ECS (Elastic Container Service) and Fargate, where containerized microservices are the standard deployment unit. Pablo brings these enterprise container patterns to Blossend's architecture, applying them selectively where containerization provides genuine value rather than defaulting to containers for every service.

Frequently Asked Questions