Vincenzo Timmel
Verified Expert in Engineering
Data Scientist and Developer
Neuenhof, Switzerland
Toptal member since July 26, 2022
Vincenzo is a data scientist with five years of professional experience. For three years, he worked for a product-data-focused eCommerce startup, in which his main areas were data cleaning and data classification using NLP tools. In this primarily remote and customer-facing position, he had to understand and implement the customer's needs. During his free time, he is a successful Kaggle and Numerai participant and loves solving complex problems.
Portfolio
Experience
- Statistics - 4 years
- Data Science - 4 years
- NumPy - 4 years
- Pandas - 4 years
- Scikit-learn - 4 years
- Python - 4 years
- Statistical Modeling - 3 years
- SciPy - 2 years
Availability
Preferred Environment
Python 3, Data Science, Statistics, NumPy, Dask, PyTorch, Statistical Modeling, Scikit-learn, Machine Learning, Pandas
The most amazing...
...product I've done is an automated machine learning model on Google Cloud for a finance tournament, Numerai, which continuously tops the scoreboard.
Work Experience
Data Analyst
ONEDOT
- Led and worked on the integration of product data into the customer's system with the help of machine learning and NLP tools, namely translation, cleaning and enhancing of product descriptions and classification of products.
- Automated a daily two-hour process by automating the fetching, categorization, and integration of around 600,000 from 4-6 files with a daily varying layout.
- Managed, engineered, tracked, and simultaneously implemented requirements for up to four data processing and integration projects.
- Led and partially implemented the solution to automate the daily integration of nearly 4.1 million products into the customer's system from around 50,000 semi-structured XML files.
Experience
Daily Auto-integration | The Product Data of 4.1 Million Articles
Image Captioning
https://github.com/kenfus/ImageCaptioningThe architecture is basically as follows:
• A pre-trained CNN model, namely ResNet50. It is used to generate features from the images.
• With the help of an embedding, the dimension is adapted to the predefined vocab size, and the embedding dimension is selected based on available computing resources.
• This vector is then passed as the first hidden state in an LSTM, and a sentence is generated from it.
Deep Learning-based Question and Answering
https://github.com/kenfus/QuestionAndAnsweringOnSquadBased on its answers, we made a recommendation to our University.
Text Sentiment Analysis
https://github.com/kenfus/SentimentClassificationFor each metric, we presented the best method and where this metric could be the most important.
Education
Bachelor's Degree in Data Science
University of Applied Sciences and Arts Northwestern Switzerland - Brugg, Switzerland
Certifications
Machine Learning
Stanford University | Via Coursera
Deep Learning
DeepLearning.AI | Via Coursera
Applied Text Mining in Python
University of Michigan | Via Coursera
Applied Machine Learning in Python
University of Michigan | Via Coursera
Skills
Libraries/APIs
NumPy, Natural Language Toolkit (NLTK), Pandas, Scikit-learn, PyTorch, SciPy, Dask
Languages
Python 3, Python, SQL
Storage
Google Cloud
Platforms
Docker
Other
Machine Learning, Data Science, Deep Learning, Data Analysis, Data Visualization, Statistics, Statistical Modeling, Natural Language Processing (NLP), User Requirements, Technical Requirements, Product Management, Data Wrangling, Time Series Analysis, Generative Pre-trained Transformers (GPT)
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