Marco Antônio Reichert Boaretto
Verified Expert in Engineering
MLOps Developer
Curitiba - State of Paraná, Brazil
Toptal member since October 4, 2024
Marco is an AI and ML engineer with eight years of experience developing end-to-end solutions. He holds a bachelor's degree in mechatronics engineering and a master's in electrical engineering, ML, and applied AI. Proficient in Python, Django, TensorFlow, MLOps, AWS, and Azure, Marco focuses on prompt engineering and multi-agent architectures for LLMs and RAG. He has expertise in fault detection, facial recognition, demand forecasting, and delivering robust and scalable solutions.
Portfolio
Experience
- Python - 8 years
- Machine Learning - 7 years
- Artificial Intelligence (AI) - 7 years
- Computer Vision - 7 years
- Hugging Face Transformers - 3 years
- Streamlit - 2 years
- LangGraph - 1 year
- LangChain - 1 year
Preferred Environment
Windows, Linux, PyCharm, Docker, GitHub
The most amazing...
...thing I've built is an AI facial recognition system for access control in a factory with 5,000 users using Python and deep learning for secure identification.
Work Experience
Senior ML Engineer
Act
- Developed a tool that enhanced agile teams by integrating an agentic retrieval-augmented generation (RAG) system, providing the product owner with real-time insights and decision-making support for backlog management.
- Designed a multi-agent architecture to automate the creation of test scenarios and code generation for multiple programming languages, improving testing efficiency and reducing manual intervention.
- Built an AI assistant for internal company use, focused on performing quality assurance (QA) on HR documents, ensuring consistency, accuracy, and compliance with internal policies and standards.
Senior ML Engineer
Robert Bosch LTDA
- Developed a facial recognition system using state-of-the-art techniques for access control in a factory plant, ensuring robust identification in uncontrolled environments and enhancing security.
- Designed a fault detection system using AI and computer vision, replacing legacy methods to prevent production errors, reducing rework costs by millions of Brazilian reais.
- Implemented a semantic search system for the tax and legislation department, enabling efficient searches across large document sets and streamlining compliance and legal research.
- Developed an automated part inspection system, replacing manual processes to reduce inspection time, minimize human error, and improve overall production quality.
- Created a predictive maintenance system that analyzes sensor data from the production line, raising alarms for potential machine part failures and allowing proactive action to prevent costly breakdowns.
ML Engineer
Fiscal Tecnologia e Automação
- Developed a free-flow system platform that detects and classifies vehicles, generates reports, and raises alarms using advanced object detection techniques like YOLOv5, ensuring real-time performance and accuracy in traffic monitoring.
- Designed and implemented a low-latency traffic light detection and red-light enforcement system for vehicles on high-speed roads, optimizing performance to ensure timely and accurate violation detection in fast-moving traffic environments.
- Created a high-accuracy license plate recognition system that reads various license plate formats across Latin America, ensuring reliable and robust performance in diverse lighting and environmental conditions.
Experience
Intelligent HR Data Analytics Dashboards
Large Language Model (LLM) agents were integrated within the analytics pipeline, enabling the analysis of complex contextual data from feedback reports and providing deeper insights into employee sentiment and satisfaction. A security layer was also implemented to ensure access control, allowing only authorized personnel to access sensitive HR data.
This project streamlined the company's HR processes, offering stakeholders faster, more accurate insights into employee trends while maintaining strict data security protocols.
Education
Master's Degree in Electrical Engineering
Federal University of Paraná - Curitba, Paraná, Brazil
Bachelor's Degree in Mechatronics Engineering
Pontifical Catholic University of Paraná - Curitba, Paraná, Brazil
Skills
Libraries/APIs
Pandas, Scikit-learn, Hugging Face Transformers, Natural Language Toolkit (NLTK)
Tools
GitHub, PyCharm, Scikit-image, Apache Airflow, AWS Command Line Interface (CLI), Plotly, AWS Glue DataBrew, NGINX
Languages
Python, C++, R, SQL
Frameworks
Django, Streamlit, Flask, LangGraph, Darknet, Spark, LlamaIndex
Paradigms
REST, Automation
Platforms
Windows, Linux, Docker, Azure, AWS IoT, Azure AI Studio
Storage
PostgreSQL, MongoDB, MySQL, Data Lakes, Redis
Other
Artificial Intelligence (AI), Machine Learning, Deep Learning, Computer Vision, APIs, FastAPI, Time Series Analysis, Forecasting, YOLOv5, Facial Recognition, LangChain, Optimization, Cryptography, Model Tuning, Technical Leadership, Quality Assurance (QA), Predictive Maintenance, Natural Language Processing (NLP), Semantic Search, Retrieval-augmented Generation (RAG), AI Agents, Semantic Kernel (SK), Amazon Bedrock AgentCore, Vector Databases, Multi Agents, OpenAI
How to Work with Toptal
Toptal matches you directly with global industry experts from our network in hours—not weeks or months.
Share your needs
Choose your talent
Start your risk-free talent trial
Top talent is in high demand.
Start hiring