Gael Eduardo Pérez Gómez

B.S. in Computer Technologies Engineering

Minor in Advanced Artificial Intelligence for Data Science

Leadership Diploma

Programming Technician

• ▸RAG in production: built the knowledge layer of an LLM assistant for a multi-tenant SaaS used by small businesses (WhatsApp and voice). Structure-aware chunking with document/section context prepended to every chunk; per-tenant knowledge base; BM25 retrieval with source attribution passed to the model; access filtering enforced twice (in SQL and at retrieval) so internal documents never reach customers. Chose lexical over vector retrieval for 30–80 chunks per tenant ($0 per query, no re-indexing on model changes), with the schema ready to move to embeddings as volume grows.
• ▸LLM security: hardened the assistant's tool-calling agent (write access) against the OWASP Top 10 for LLM Applications — output filtering, URL allowlist, 57 automated security tests.
• ▸Same platform: Node.js/PostgreSQL with payments (Conekta), CFDI 4.0 e-invoicing and scheduling; 2,700+ automated tests including cross-tenant isolation; versioned SQL migrations; alerting and health checks.
• ▸Art studio platform (in production): Next.js 15, Node/Express, PostgreSQL, WhatsApp booking bot, S3 gallery, admin panel.
• ▸Python ETL automation for SME clients, raising data accuracy above 95% and saving 20+ hours/month; integrations with Twilio, payment gateways and AWS S3.

• ▸Designed and implemented key microservice components for the cloud migration of a legacy Oracle-based system to AWS, applying Hexagonal Architecture to improve scalability, maintainability, and domain isolation.
• ▸Implemented AWS Lambda functions for critical business flows, including authentication, login, and password recovery.
• ▸Designed and integrated Amazon SQS for asynchronous, event-driven communication between services.
• ▸Implemented Amazon SES for transactional and notification email delivery.
• ▸Defined and documented service contracts, architecture decisions, and system workflows during the migration process.
• ▸Supported the lead architect in planning, documentation, and auditing of services to ensure a controlled and low-risk migration.
• ▸Implemented AI Development Agents using Kiro, reducing overall migration time by approximately 60% while improving code consistency and delivery speed.
• ▸Collaborated closely with business and engineering teams to define modernization strategies and reduce technical debt.

• ▸Built the backend and frontend of a strategic decision-making dashboard used by a Fortune 500 US retail client to analyze and reduce inventory shrinkage (merchandise loss) across its operations.

• ▸Modernized legacy Java systems into AWS Lambda microservices, reducing maintenance costs by 30% and improving scalability.
• ▸Migrated Angular/Java MVC apps to React and TypeScript, cutting UI bug reports by 40%.
• ▸Streamlined CI/CD pipelines with CircleCI and Jenkins, reducing deployment time by 50%.
• ▸Implemented Swagger/OpenAPI for API documentation, reducing developer onboarding time.
• ▸Enhanced testing with Jest, Pytest, Mocha, Selenium, decreasing production bugs by 25%.

• ▸Migrated monolithic systems to microservices (Python, Java, Node.js), enabling independent scaling of features.
• ▸Automated deployments with Jenkins and CircleCI, reducing release times by 50%.
• ▸Containerized apps with Docker, reducing deployment errors by 35%.
• ▸Improved API performance with PostgreSQL and MongoDB, reducing query times by 30%.
• ▸Enhanced UI responsiveness with React and TypeScript, increasing user satisfaction by 25%.

• ▸Delivered new product features end to end in JavaScript/TypeScript across frontend and backend for a US automotive services & marketplace client.
• ▸Diagnosed, traced and fixed production bugs using Jenkins pipelines and AWS logs; worked through the full ticket lifecycle with QA.

• ▸Built and maintained the company CRM and client-acquisition platform: Python and Node.js services on AWS EC2/S3 with Docker.
• ▸Created REST APIs documented with Swagger, and automated tests with Pytest.
• ▸Built the landing page and onboarding flow with React and Flask.