Igor Souza

Python Engineer Middle

• Built a Python system for network field operations, communicating with routers over Telnet and SSH to manage and monitor them; also helped on the JavaScript frontend and Java/Hibernate backend.
• Moved to a sibling project rebuilding the same domain on microservices with Kafka as the event backbone, developing the service that routed UI requests to routers.
• Helped design the architecture extending router communication from Telnet/SSH to REST APIs, and managed Kubernetes Helm charts and Grafana dashboards for observability.
• Led the Linux distro upgrade for internal infrastructure, resolving GRUB failures and kernel panics to ensure a smooth migration.

Tech: Python · Linux · Networking · Java · Perl

Python Engineer Middle

• Owned a Kubernetes microservice retrieving bills from Brazil's national bill service and storing them for downstream use.
• Tracked down a bill-loss bug between the microservice and the legacy Django monolith by debugging inside Kubernetes containers, inspecting RabbitMQ message flow, and learning Ruby to audit a connector; fixed it on both sides.
• Built a shared messaging package adopted by other Python microservices.
• Helped build the Django API-based microservice that replaced the legacy monolith.

Tech: Python · Flask · Docker · Kubernetes · RabbitMQ · GCP

Python Engineer Junior

• Built the backend for an assistive browser, powering two in-house models: one identifying icons and outputting their meaning as words, another extracting text and context from images.
• Designed the pipeline end-to-end with Flask behind an AWS Lambda function, using Docker and Docker Compose to emulate Lambda locally; also contributed to the Vue.js browser frontend.
• Built the Flask backend and React frontend for a dashboard plugin integrated into QA program managed by Microsoft Products, showing QA metrics and forecasts, and helped deliver the forecasting model.
• Partnered closely with the team architect on key decisions, tackling model response-time optimization and Azure plugin-approval requirements, in a team mostly composed of trainees.

Tech: TensorFlow · Torch · Vue · React · Streamlit · Docker · Azure · AWS

Python Engineer Junior

• Built and trained a computer vision model to detect and extract text from video frames, storing captured data for downstream use.
• Trained a model to extract structured information from photos of bills, and built the Flask backend that received the image, ran inference and returned the parsed data.
• Managed the image datasets used for training, including a Qt desktop tool built to capture and label training images.
• Proficient use of Docker, Linux and SSH for development and deployment.

Tech: Python · TensorFlow · OpenCV · Flask · Docker · Linux

Bachelor in Software Engineering

• NLP - Natural Language Processing
• REST Api com Python e Flask
• Docker and Kubernetes: The Complete Guide
• Artificial Intelligence: Reinforcement Learning in Python