Javier – RAG, LLM, Python

A cloud-native logistics platform for modernizing legacy systems through scalable microservices, Kubernetes-based deployments, event-driven shipment processing, and RAG-powered semantic retrieval.

• Contributed to the migration of legacy on-premise features into cloud-native microservices deployed on Google Kubernetes Engine (GKE), helping improve scalability and operational reliability;
• Collaborated closely with Product and integration teams to refine requirements and ensure successful end-to-end feature delivery from development to pre-production environments;
• Participated in architectural discussions to align new services with distributed systems best practices and internal cloud guidelines;
• Stored and indexed embeddings in Elasticsearch vector search to support semantic retrieval as part of a Retrieval-Augmented Generation (RAG) workflow, enabling relevant context to be retrieved and provided to LLMs;
• Implemented high-performance backend services using Java 21, JAX-RS, JPA/Hibernate, and WebSphere Liberty, ensuring clean design and maintainability;
• Designed event-driven processing for shipment lifecycle events using Pub/Sub;
• Contributed to CI/CD processes using TeamCity and Octopus, supporting reliable deployments to GCP-based Kubernetes environments;
• Generated semantic embeddings using Python-based services and managed batch processing for large volumes of enterprise data;
• Ensured high-throughput, idempotent, and scalable EDI status delivery.
• Investigated and resolved production issues, focusing on performance optimization and system stability;
• Worked within Scrum-based agile teams, participating in planning, refinement, and cross-team synchronization meetings to ensure alignment with business objectives.

An enterprise backend platform built on a scalable microservices architecture, combining event-driven communication, AWS cloud infrastructure, and AI-powered document intelligence through RAG workflows and LLM-based agents.

• Participated in system design discussions and translated product requirements into scalable backend solutions, ensuring alignment through proper technical documentation and implementation guidelines;
• Designed and implemented microservices following Clean Architecture and Domain-Driven Design (DDD) principles;
• Addressed a race condition scenario in a third-party authentication flow by introducing Kafka as a distributed broker to serialize requests and Redis as a caching layer to improve consistency and performance;
• Implemented event-driven communication using webhooks in a distributed microservices architecture to handle asynchronous client responses;
• Designed a scalable processing strategy for a microservice by leveraging AWS SQS, Lambda functions, and EventBridge to dynamically schedule and control workload processing;
• Added observability to microservices by integrating Micrometer (MeterRegistry) with AWS CloudWatch for metrics collection and monitoring;
• Implemented Retrieval-Augmented Generation (RAG) architectures using embeddings and vector storage to enable contextual data access across internal systems;
• Contributed to DevOps processes, including CloudFormation templates and Jenkins pipeline configuration for CI/CD automation;
• Built a document processing service using Quarkus and Mutiny to handle OCR workflows, exposing metrics through Micrometer and CloudWatch;
• Designed and implemented an MCP Server to enable intelligent document management for semantic indexing (embeddings) and contextual retrieval capabilities, allowing LLM-based agents to query, reason, and interact with enterprise documents in a secure and scalable manner;
• Ensured performance considerations and DevOps best practices were incorporated throughout the development lifecycle;
• Worked within Scrum-based agile teams, contributing to iterative delivery and cross-functional collaboration.

A cloud-native backend platform focused on scalable architecture, secure payment processing, and reliable transaction handling through event-driven communication, serverless infrastructure, and containerized deployments.

• Defined backend architecture strategy leveraging Clean Architecture to decouple domain logic from infrastructure concerns;
• Designed event-driven communication patterns using AWS SNS/SQS and serverless compute with Lambda to improve scalability and cost-efficiency;
• Led the integration of external payment services (Stripe), ensuring secure transaction handling and fault tolerance;
• Modernized legacy modules and aligned them with cloud-native and containerized deployment standards;
• Established deployment workflows across environments and ensured production-grade reliability;
• Acted as a technical reference for architectural decisions and implementation patterns.

A cloud-native time and attendance platform for workforce scheduling and tracking, built with microservices and event-driven architecture to support large-scale operations and high-volume data processing.

• Contributed to architectural direction of a cloud-native, event-driven platform leveraging Pub/Sub and serverless patterns to support high-throughput processing;
• Designed scalable microservices using Event Sourcing and DDD principles, enabling reliable state reconstruction and domain consistency;
• Led migration of legacy scheduling systems to Kubernetes and GCP-based serverless workloads, handling 500K+ records in minutes through optimized concurrency strategies;
• Defined horizontal scaling strategies (Kubernetes RPS-based autoscaling and event-driven Lambda orchestration);
• Established distributed system best practices including JWT asymmetric authentication, observability (Splunk), and multithreaded processing standards;
• Mentored engineers and influenced cross-team architectural decisions in a microservices ecosystem.