Wington Brito
Full-Stack Machine Learning Engineer with 10+ years of experience driving digital transformation and engineering operations for up to 10K+ concurrent users. Skilled in designing high-performance systems, modeling APIs, and automating SaaS infrastructures to reduce operational overhead by 60%. Leveraged advanced machine learning solutions to boost lead conversion by 30% and accelerate development velocity by 40%. A strong communicator and mentor, I align technical vision with business goals to deliver measurable outcomes and strategic growth.
REVIEWS FROM CLIENTS
• Engineered high-performance backend systems (Node.js, Angular/React, PostgreSQL) for 10K+ concurrent users.
• Designed distrib...
• Engineered high-performance backend systems (Node.js, Angular/React, PostgreSQL) for 10K+ concurrent users.
• Designed distributed microservices with AWS Lambda, S3, and Kinesis—reducing latency by 40%.
• Built API-driven architectures ensuring scalability and security.
• Led cross-functional teams, mentoring junior developers and optimizing CI/CD workflows.
• Directed end-to-end ML model development for KooperAI; designed AI pipelines (Flux, GPT-3.5-Turbo, BGE-Large-En-V1.5) increasing engagement by 35%.
• Architected scalable data pipelines for real-time and batch AI inference.
• Managed a 6-person tech team with actionable documentation and hands-on support—boosting development velocity by 40%.
• Developed a dynamic quiz generator ML pipeline leveraging GPT-3 models and cloud-native services.
• Enhanced recommendation pipelines by implementing a vector-based database using Cloudflare’s Vectorize framework.
• Implemented a retry API mechanism for AWS Lambdas covering 10+ functions with comprehensive testing.
• Developed an end-to-end ML pipeline for social media lead scoring—boosting conversion by 30%.
• Engineered real-time data pipe...
• Developed an end-to-end ML pipeline for social media lead scoring—boosting conversion by 30%.
• Engineered real-time data pipelines with AWS Kinesis, Kafka, and Spark Streaming—reducing model drift by 40%.
• Built anonymized sentiment analysis models with TensorFlow and PyTorch, improving prediction accuracy by 20%.
• Automated big data preprocessing, feature engineering, and training with Apache Airflow and Dataflow—cutting training times by 60%.
• Streamlined large-scale data ingestion for BigQuery and Airflow, improving scalability.
• Mentored engineers in MLOps and optimized DAG configurations.
- Developed consumer-facing applications serving over 25,880 users, improving overall user experience by 40%.
- Developed consumer-facing applications serving over 25,880 users, improving overall user experience by 40%.