About
Not Playing
–Spotify
João Pedro
Data Scientist & AI Engineer
+55(92)99104-4363Manaus, Amazonas, 🇧🇷 🐆
Hi there
I'm João Pedro Pereira Santiago — Data Scientist and AI Engineer building production ML systems from the ground up. Based in Manaus, Brazil, working globally.
I go from raw data to deployed model. My focus is on MLOps pipelines, computer vision, deep learning, and LLM integration — systems that run in production, not just notebooks.
I've shipped solutions across industrial energy analytics, assistive AI, and business intelligence — working with GCP, real-time IoT streams, and multi-disciplinary teams.
📍 Based in Brazil, working globally
Career
Resume
- Design and maintain scalable ML systems on Google Cloud Platform, covering the full model lifecycle from experimentation to production.
- Architect end-to-end AI pipelines with production-grade MLOps: CI/CD, model versioning, automated retraining, and canary deployments.
- Integrate and fine-tune LLMs and VLMs for domain-specific enterprise applications.
- Build internal observability and model governance frameworks for systematic monitoring of drift, performance degradation, and data quality.
- Collaborate with multidisciplinary teams of data engineers, software engineers, and domain experts to deliver reliable AI solutions at scale.
- Lead development of a deep learning system for real-time Brazilian Sign Language (Libras) translation into natural language text.
- Designed keypoint extraction pipelines using MediaPipe; trained sequence classification models with PyTorch and TensorFlow.
- Curated and preprocessed large-scale gesture recognition datasets for robust training under distribution variability.
- Developed computer vision systems enabling real-time obstacle detection and scene understanding for visually impaired users.
- Implemented deep learning object detection pipelines using YOLO and OpenCV, optimized for low-latency inference.
- Applied model compression techniques (pruning, quantization) for deployment on resource-constrained embedded devices.
- Designed and deployed ML systems for large-scale industrial energy analytics, processing real-time IoT sensor streams.
- Engineered end-to-end data pipelines orchestrated with Dagster: ingestion, transformation, feature engineering, and model training.
- Built forecasting, anomaly detection, and pattern recognition models to optimize industrial energy consumption.
- Designed scalable microservice backends exposing trained ML models through performance-optimized REST APIs.
- Implemented full MLOps lifecycle: automated CI/CD, production monitoring, and model performance tracking.
- Designed scalable data platforms integrating AI-powered automation for business intelligence workflows.
- Built high-performance ETL pipelines and intelligent data extraction systems processing structured and semi-structured data at scale.
- Developed REST and GraphQL APIs using FastAPI and Django, following clean architecture and security-first principles.
- Deployed containerized microservices with Docker and Kubernetes, ensuring reproducibility and horizontal scalability.
- Built interactive dashboards and visualization systems using React and Next.js for operational decision support.
- Building theoretical and practical foundations in software engineering, algorithms, and systems design.
- Project-based curriculum integrating academic research with real-world development.
- First university program at Escola Superior de Tecnologia — not completed.
- Where I first learned to program in C++. Pivoted from naval engineering into software development — and never looked back.
Tech stack
Machine Learning
Backend & Web
Site built with Next.js and Tailwind CSS
Certifications
Machine Learning Specialization
Stanford University · Coursera
Unsupervised Learning, Recommenders & Reinforcement Learning
Stanford University · Coursera
Supervised Machine Learning: Regression & Classification
Stanford University · Coursera
Advanced Learning Algorithms
Stanford University · Coursera
LFS116: PyTorch and Deep Learning for Decision Makers
Linux Foundation
LFS120: Conversational AI — Ensuring Compliance and Mitigating Risks
Linux Foundation
AI Agents Fundamentals
Hugging Face
Time Series Forecasting
Samsung Ocean
Introduction to Interactive Language Models and Chatbots
Samsung Ocean
Deep Learning Introduction
Samsung Ocean
EF SET English Certificate (B1 Intermediate)
EF SET