Software Engineer Senior
Resume:
Professional Summary
Senior Software Engineer with 8+ years of hands-on experience leading global teams and delivering production-ready AI/ML, backend, and cloud solutions. Highly skilled at building scalable systems, driving automation, and translating complex business needs into powerful, measurable results. Passionate mentor and technical leader, recognized for optimizing ML workflows, cutting operational costs, and enabling high- growth teams to ship faster and more reliably.
Skills
- Programming: Python (8+ yrs), Go (3+ yrs), Node.js, FastAPI, Bash, TypeScript
- AI/ML & Data Engineering: Kubeflow, MLflow, Airflow, Spark, PyTorch, TensorFlow, Scikit-learn, Transformers, NLP, RL agents, Model Monitoring
- Cloud & DevOps: GCP, AWS, Azure, Docker, Kubernetes, Terraform, Jenkins, GitHub Actions, Prometheus, Grafana
- Web & Visualization: React.js, D3.js, Tableau, Vue.js
- Blockchain: Ethereum, Solidity, Hyperledger
- Databases: PostgreSQL, Cassandra, Redis, Neo4j
- Edge/IoT: Nvidia Jetson, ESP32, MQTT
- Collaboration: Agile, Mentoring, Technical Writing, Team Leadership Professional Experience
Zapier - Senior MLOps & AI Engineer
San Francisco, CA, USA, Remote (May 2023 - Apr 2025)
- Built and maintained a robust MLOps platform for 22 teams, leveraging Kubeflow, MLflow, and Docker to cut model deployment and retraining cycles by 60%, supporting over 500 million API calls per month.
- Automated resource scaling and cloud cost management in GCP with Terraform and Kubernetes, saving $52,000 annually while maintaining 99.99% uptime across 1,200+ production containers.
- Integrated Prometheus and Grafana to deliver real-time observability, reducing incident response time from 90 minutes to 54 minutes on average, and increasing SLA compliance to 99.95%.
- Migrated 70+ legacy ETL and batch workflows to Apache Airflow and PySpark, doubling data pipeline throughput from 1.2TB to 2.4TB per day and improving analytics data freshness by 40%.
- Developed and maintained CI/CD automation with GitHub Actions and Jenkins, increasing software release frequency from 2 to 6 per week and reducing rollback rates by 43%.
- Implemented AI model monitoring and drift detection, cutting production model errors by 30% and boosting customer satisfaction scores by 15% in post-release surveys.
- Mentored and led a team of 7 engineers, increasing team delivery velocity by 35% and reducing onboarding time for new hires by 50% through structured code reviews and training sessions.
- Collaborated directly with product and business teams to align AI initiatives with customer needs, driving a 30% increase in feature adoption and generating $1.2 million in additional annual revenue.
- Enhanced data lineage and compliance processes, ensuring 100% auditability and passing all internal and external security reviews without a single finding.
Joas Carvalho
Senior Software Engineer MLOps,
AI/ML, Cloud, Blockchain
Tibau do Sul, RN, Brazil
**************@*****.***
https://www.linkedin.com/in/joas-carvalho-036472363 https://github.com/wilfrid51
Vianai Systems (vian.ai) - Senior MLOps Engineer
Palo Alto, California, USA, Remote (Jan 2021 - Apr 2023)
- Automated the ML lifecycle for 15+ predictive models on AWS and GCP, reducing retraining costs by 38% and saving
$180,000 per year across 4 product lines.
- Designed and managed Spark and Airflow ETL pipelines that scaled to process 3.5 billion records monthly, increasing throughput by 50% and supporting real-time analytics for over 1.4 million healthcare records.
- Integrated explainable AI modules, raising stakeholder adoption by 42% and accelerating regulatory approval for new models by 4 months.
- Migrated model serving infrastructure to Vertex AI, reducing operational overhead by 28% and freeing $400,000 for new R&D projects.
- Built and deployed Kafka-based streaming ingestion for healthcare event processing, increasing data velocity by 55% and enabling 24/7 diagnostic capabilities for 80+ hospital sites.
- Led internal MLOps bootcamps, training 33 engineers in modern cloud-native tooling and boosting overall team productivity by 28%.
eMed - Machine Learning Engineer
London, UK, Remote (Jun 2019 - Dec 2020)
- Architected and deployed containerized ML services on Kubernetes, supporting over 200,000 daily patient risk predictions and achieving 99.99% uptime across 12 clinical sites.
- Automated ML pipelines on AWS SageMaker and GCP Vertex AI, reducing manual retraining hours by 63% and cutting model deployment time from 6 days to just 2 days.
- Launched LLM-based medical summarization tools, increasing clinical documentation efficiency by 16% and reducing error rates by 13% as measured in quarterly quality reviews.
- Established advanced monitoring and alerting with Grafana and ELK, cutting incident response time by 33% and ensuring uninterrupted service for 1.1 million patients.
- Created automated model validation and rollback, reducing production incidents by 38% and minimizing downtime to less than 20 minutes per event.
Zalando - Machine Learning Engineer
Berlin, Germany, Remote, (Jul 2017 - May 2019)
- Developed and deployed deep learning models for personalized recommendations, increasing conversion rates by 12% and generating €7 million in incremental annual revenue across 23 European markets.
- Automated Airflow and BigQuery data pipelines, reducing data latency by 45% and enabling personalized marketing for over 20 million users.
- Optimized hyperparameter tuning, reducing training costs by 30% and saving 1,700 compute hours per year, directly lowering cloud spending by €110,000 annually.
- Engineered high-availability ML APIs with 99.98% uptime, supporting seamless integration with 6 partner systems and driving 18% faster feature rollouts.
- Enhanced drift monitoring, cutting false positives by 56% and improving customer experience scores by 11% in post- purchase surveys.
TOTVS - Backend & Data Engineer
São Paulo, Brazil, Remote (May 2015 - Jun 2017)
- Designed and delivered ETL solutions for 25+ enterprise clients, improving pipeline reliability and reducing latency by 22%, processing over 4.8TB of data monthly.
- Introduced automated data quality checks, decreasing production errors by 32% and improving analytics accuracy, leading to a 12% reduction in client support tickets.
- Supported migration of 18 legacy systems to Azure cloud, reducing operational costs by 18% and increasing system scalability and availability to 99.96%.
- Developed and maintained backend APIs used by 2.5 million monthly users and 60,000+ practitioners, enabling 24/7 access to critical business services.
- Mentored 9 junior developers, leading code quality initiatives and boosting team productivity by 27% over two years. Education
State University of Campinas
Master of Science, (2015 - 2017)
- Thesis: “Federated Learning and Secure Multi-Party Computation for Medical Imaging”
- Research Assistant: AI-powered robotics, quantum algorithm prototyping Federal University of Rio Grande do Norte
Bachelor of Science, (2010 - 2014)
- Capstone: “Blockchain-based Identity Management for Smart Cities”
- Hackathon Winner: IoT and Edge Computing Challenge
Contact this candidate