Enhancing Online Education - Python and Data Science Instructor
2023 - Present
Challenge
Working with one of the largest online education providers in Brazil, Hashtag Treinamentos, with over 80,000 students trained in their courses, the challenge was to develop video and written materials for Python and Data Science courses to ensure a high-quality learning experience.

Solution
Upon embracing this pivotal role, my initial endeavor was to thoroughly understand the learning objectives and expectations of both the institution and the students. This entailed engaging in discussions with the curriculum development team and analyzing feedback from previous courses. With this foundational understanding, I commenced the development and rigorous review of teaching materials. This meticulous process involved not only crafting the content but ensuring it was pedagogically sound, engaging, and aligned with the latest industry standards.
A significant portion of the teaching materials was dedicated to covering core Python libraries indispensable for data science pursuits, such as NumPy, Pandas, Matplotlib, Seaborn, Scikit-Learn, TensorFlow, PyTorch, and Plotly. The goal was to demystify complex concepts and provide a hands-on approach to learning through practical examples and real-world applications.
To augment the learning experience further, I produced and narrated a series of educational videos on Python. These videos were meticulously designed to break down complex topics into digestible segments, and were disseminated not only on the company’s e-learning platform but also on various social media channels, expanding the reach and impact of the educational content.
Results
One of the hallmark initiatives was the integration of real-world projects into the curriculum. This was driven by the aspiration to bridge the gap between theoretical knowledge and practical market scenarios. By doing so, students were endowed with a pragmatic understanding and experience of industry-relevant challenges and solutions, which is quintessential for their career readiness and success.
Furthermore, the materials I engineered proved to be instrumental in facilitating the learning journey for thousands of students on the company’s educational platform. This significant contribution resonates with the company’s overarching mission of delivering high-quality education and has left a lasting impact on the educational trajectory of many aspiring data scientists and programmers.
Throughout this journey, I continually fostered a stimulating learning environment. By providing continuous support, feedback, and additional resources to students, I ensured their success and comprehension of complex topics. The open channel of communication also allowed for the iterative improvement of the course materials, thereby elevating the overall learning experience.
The endeavor at Hashtag Treinamentos not only honed my skills in curriculum development and educational content creation but also enriched my understanding of the diverse learning needs and aspirations of students in the realm of Python and Data Science.
Project Showcase
In my pursuit to provide a practical, hands-on learning experience, I’ve initiated a section dedicated to showcasing the real-world projects integrated into the curriculum. These projects are meticulously designed to emulate real-world scenarios, ensuring students gain pragmatic insights and hands-on experience essential for the modern job market.
Optimizing Marketing Campaigns Project
This project is rooted in a real recruitment challenge posed by iFood, a notable company in Brazil, simulating a scenario from a well-established retail food company. It encompasses a comprehensive exploratory analysis, customer segmentation, and a classification model to predict customer responses to marketing campaigns. The project is structured to mimic a real-world data analyst selection process, offering students a pragmatic insight into data preprocessing, exploratory analysis, and presenting findings to both technical and non-technical audiences. The repository is structured with data, images, notebooks, and reports folders, illustrating a well-organized data science project workflow. Through this project, students also learn good programming practices in Python, utilizing libraries like Scikit-Learn for building and optimizing models.
Credit Card Fraud Detection Project
This project is centered around the detection of fraudulent credit card transactions, which is a critical challenge faced by financial institutions worldwide. The project involves data preprocessing, exploratory data analysis, and the development of a classification model to identify fraudulent transactions. The dataset used in this project is highly imbalanced, reflecting the real-world scenario of credit card fraud detection. By working on this project, students gain a profound understanding of handling imbalanced datasets, feature engineering, and model evaluation techniques. The project is structured with a detailed notebooks and reports, guiding students through the entire data science project lifecycle.
Customer Churn Prediction Project
This project revolves around predicting customer churn, a critical challenge faced by companies across various industries. The project entails data preprocessing, exploratory data analysis, and the development of a classification model to predict customer churn. The dataset used in this project is sourced from a telecommunications company, providing students with a real-world scenario to work on. By engaging with this project, students gain practical experience dealing with categorical variables and their encoding, feature scaling, and model evaluation techniques. The project is structured with notebooks and reports, guiding students through the entire data science project lifecycle.
Some Lecture Notes
While the projects above are designed to provide a hands-on learning experience, the following lecture notes are aimed at enhancing the theoretical understanding of key concepts in Machine Learning and Data Science.
- Metrics for Binary Classification with Imbalanced Datasets
- Understanding Logistic Regression Coefficients
Stay tuned as more projects are in the pipeline, aimed at further bridging the theoretical-practical gap, and will be shared here once finalized.