Rio de Janeiro, Brazil
As a Lead AI Engineer and Technical Architect based in Brazil, I specialize in bridging the profound gap between cutting-edge artificial intelligence research and highly scalable, production-ready enterprise systems. With extensive engineering experience and a specialized focus on the modern AI stack, I direct cross-functional, global teams to architect, deploy, and operationalize Generative AI, Large Language Models (LLMs), and autonomous Agentic AI workflows for top-tier international clients, including the Microsoft ecosystem. My engineering leadership ethos is built on the rigorous translation of complex deep learning constraints into strategic, measurable business value. I excel in synchronous, nearshore collaboration with North American stakeholders, ensuring rapid time-to-value without compromising architectural integrity, data governance, or system security. Rather than stopping at exploratory modeling within isolated environments, I take definitive ownership of the entire production lifecycle—from foundational model evaluation and selection to cloud-native, fault-tolerant deployment. Technical Expertise & Production Arsenal: ➤ Advanced AI Architectures: Deep Neural Networks, Transformers, Large Language Models (OpenAI, Claude, Gemini), Multimodal AI, and advanced Computer Vision systems. ➤ Agentic AI & Complex Workflows: Designing autonomous multi-agent systems, executing Tool-Use/Function Calling patterns, and engineering intelligent semantic routing using industry-leading frameworks including LangGraph, AutoGen, CrewAI, DeepEval. ➤ Production Agentic Development / Deep Learning: Enterprise Retrieval-Augmented Generation (RAG), dynamic Fine-Tuning, Low-Rank Adaptation (LoRA), Post-Training Quantization (PTQ), and deploying high-throughput models leveraging PyTorch. ➤ MLOps & Scalable Cloud Infrastructure: End-to-end model lifecycle management across AWS, Azure, and Google Cloud Platform (GCP). Deep expertise in container orchestration (Kubernetes, Docker), continuous integration pipelines (CI/CD for ML), rigorous Model Monitoring (detecting data drift and latency degradation), and building resilient REST APIs for high-concurrency AI agent serving. Backed by comprehensive industry certifications across AWS and Microsoft, including the AWS Certified Machine Learning Specialty and the GitHub Copilot (GH-300) certification (+ 5 other industry certifications), I build secure, highly available AI infrastructures that drive industrial evolution. I am deeply committed to mentoring junior data scientists, fostering agile engineering cultures.
Leading cross-functional, global engineering teams to architect, engineer, and deploy enterprise-grade Agentic AI and Generative AI solutions for the Microsoft ecosystem, bridging advanced deep learning capabilities with highly scalable cloud infrastructure. • Technical Leadership & Architectural Direction: Spearheaded the engineering delivery of complex, concurrent AI initiatives, managing and continuously mentoring a distributed team of data scientists and software engineers. Maintained definitive ownership of the technical roadmap, complex architectural decision-making, and the critical alignment between offshore engineering outputs and the strategic expectations of North American stakeholders. • Agentic AI Orchestration & Deployment: Architected and deployed sophisticated autonomous multi-agent workflows utilizing advanced Large Language Models. Engineered complex reasoning pipelines, autonomous tool-use patterns, and dynamic context routing, which significantly enhanced internal marketing efficiency and automated complex, multi-step business processes across the enterprise. • Production-Ready Deep Learning: Transitioned exploratory generative AI models into robust, fault-tolerant production systems. Led the architectural design of highly concurrent APIs for AI Agent serving, ensuring low-latency inference, optimal compute utilization, and high availability within a rigorous, enterprise-grade Azure cloud environment.
Acted as the lead technical consultant for advanced Intelligent Document Processing (IDP) and data lake initiatives, leveraging the comprehensive AWS cloud ecosystem to deliver highly scalable computer vision and generative AI architectures for enterprise clients. • Intelligent Document Processing (IDP) Architecture: Engineered and deployed robust Generative AI classification models and automated extraction pipelines. Seamlessly integrated Large Language Models (LLMs) with advanced Computer Vision techniques to process highly diverse, unstructured document formats. This architectural implementation drastically reduced manual intervention requirements and significantly increased Straight-Through Processing (STP) rates for downstream business units. • AWS Cloud Infrastructure Mastery: Designed, deployed, and managed complex end-to-end Machine Learning workflows utilizing AWS native services, prominently including Amazon SageMaker, Bedrock, and Lambda. Built highly scalable, serverless data architectures capable of ingesting and analyzing high-volume document streams while maintaining optimal cloud compute cost-efficiency. • Agentic Workflow Implementation: Pioneered the integration of early-stage Agentic AI workflows designed for complex document segmentation and semantic search. This allowed the system to autonomously route, cryptographically validate, and structure extracted textual data directly into centralized downstream enterprise data lakes. • Deep Learning Optimization & Evaluation: Continuously evaluated and experimented with state-of-the-art foundation models to ensure technological superiority. Applied rigorous prompt engineering, dynamic fine-tuning, and performance optimization techniques to maximize extraction accuracy while actively minimizing Character Error Rates (CER) and Word Error Rates (WER) during high-throughput inference cycles.
Architected secure, strictly compliant Generative AI solutions and intelligent Virtual Assistants for one of Latin America’s largest financial institutions, deploying sophisticated natural language systems to democratize secure data access across highly regulated enterprise departments. • Enterprise RAG Architectures: Developed and deployed advanced Retrieval-Augmented Generation (RAG) pipelines designed to support institutional banking analysts and senior management. Utilized complex LangChain frameworks, high-dimensional vector embeddings, and dynamic few-shot learning algorithms to ensure all LLM outputs were strictly grounded in approved internal financial policies and regulatory documentation. • Financial Virtual Assistants & Governance: Engineered specialized, context-aware AI conversational agents utilized extensively by legal, finance, and communications departments. Implemented rigorous architectural guardrails, semantic routing protocols, and active hallucination-mitigation systems to ensure absolute mathematical accuracy and regulatory compliance in all financial and KPI-related querying. • Cloud ML Deployment & Orchestration: Led the critical transition of machine learning models from local, exploratory development environments to highly secure production states utilizing the Google Cloud Platform (GCP). Leveraged Vertex AI and Google BigQuery to manage highly scalable training pipelines, complex feature engineering, and strict model versioning protocols for proprietary foundation models. • Business Intelligence & Strategic Integration: Bridged the vital gap between complex algorithmic outputs and overarching banking strategy. Translated intricate AI concepts and model performance data into operationally feasible tools, utilizing platforms like Power BI to visualize system efficacy, monitor user adoption metrics, and generate actionable analytical insights for executive banking stakeholders.
- Responsible for the intelligent implementation of AI and conversational LLMs in systems. - Development and integration of Bots through the OpenAI API, Gemini, and others. - Shared my knowledge in various segments of Data Science with the team.
• Solid, full-time experience with the planning, execution, and analysis of Load Tests, Stress Tests, and their variations. • Knowledge in real-time analysis of important reports during the execution of Performance Tests, such as: Error Reports, Throughput Rate Graphs, Response Time Graphs, and others. • Experience in creating and reviewing various reports, as well as the storytelling skills necessary to perform these tasks. • Commitment to delivering exceptional service to customers in order to ensure long-term loyality. • Acquired broad skills in active listening and analytical thinking.