Principal Machine Learning Engineer

Building production-grade ML and Generative AI systems that deliver measurable business impact.

4x Google Cloud Certified and 2x AWS Certified professional, specialized in Python, MLOps, LLMs, RAG, and multi-agent architectures across AWS, GCP, and Azure.

LinkedIn GitHub

Brazil
Open to Data Scientist, ML Engineer, Python Developer, and Software Engineer roles

4x

Google Cloud Certifications

2x

AWS Certifications

500+

Professional Connections

Top 5,000

AWS GenAI Developer Professional Early Adopter

About

Engineer profile

I am Luis Henrique Miranda Queiroz, a Principal Machine Learning Engineer at Marlabs and a Computer Science undergraduate at the Federal University of Piaui (UFPI).

My work is focused on AI systems that move from prototype to production with reliability, governance, and business value. I combine ML engineering, MLOps, and cloud architecture to deliver robust solutions in real-world environments.

Key capabilities

End-to-end ML lifecycle: data, training, deployment, monitoring, and retraining
Generative AI applications with RAG, hybrid search, and multi-agent orchestration
Production-grade cloud implementations on AWS, GCP, and Azure
Applied NLP for enterprise and public-sector use cases

Impact

Selected outcomes

AWS Resume-to-Job Matching System

Architected and implemented an intelligent matching platform using RAG, Pinecone-based hybrid search, metadata filters, and cross-encoder reranking.

Result: Faster screening cycles, improved recommendation quality, and direct contribution to roughly 50% revenue growth.

MLOps Pipeline for Oncology Risk Prediction

Designed a complete SageMaker pipeline with RDS ingestion, LightGBM training, versioning, GenAI explainability, and real-time endpoints.

Result: Robust production governance with drift detection and automated retraining triggers.

AI Agent with Bedrock AgentCore + MCP

Built an intelligent candidate-support agent with contextual responses, scalable architecture, and full observability via Langfuse.

Result: Seamless internal integration and support for hundreds of concurrent users.

Public Security Multimodal Chatbot

Co-developed an LLM-powered multimodal chatbot for police incident reporting, integrated with WhatsApp and cloud infrastructure.

Result: Research project evolved into BO Facil, an operational public service in Piaui.

Experience

Professional journey

Mar 2026 - Present

Principal Machine Learning Engineer | Marlabs

Remote from Brazil, leading high-impact ML and GenAI initiatives.

Jun 2025 - Mar 2026

Data Scientist | SOUTH SYSTEM

Led architecture and implementation of ML and GenAI systems for internal products and enterprise clients.

Sep 2024 - Jun 2025

Machine Learning Engineer | SantoDigital

Developed GCP-based AI and MLOps solutions, including multi-agent systems and model fine-tuning workflows.

Jan 2024 - Sep 2024

Machine Learning Engineer Intern | SantoDigital

Built CI/CD retraining pipelines and GenAI chatbots with RAG and function-calling patterns.

Sep 2024 - Sep 2025

Scientific Researcher | CNPq / SSP-PI Project

Developed NLP solutions for police report analysis and chatbot-assisted public service workflows.

Credentials

Certifications and recognition

Featured Certifications

Publications and awards

Paper published at ENUCOMPI 2025: Multimodal Chatbot with LLMs for Incident Reporting in Public Security Services
1st place at SIUFPI 2025 with applied research in public security AI
Bronze Medal - Brazilian Physics Olympiad (3rd Stage)

Projects