Bilal Mahmood
Verified Expert in Engineering
Data Scientist and Machine Learning Developer
Bergen, Norway
Toptal member since October 5, 2022
Bilal is an experienced data scientist with solid knowledge of machine learning, data analysis, and visualization. He offers four years of professional experience in data science, working on projects related to building and deploying machine learning models, extracting insights from data, and managing data. Bilal is passionate about working with teams whose vision aligns with his values and specializations.
Portfolio
Experience
Availability
Preferred Environment
Ubuntu, Visual Studio Code (VS Code), Jupyter Notebook, Slack, Anaconda, Python, Gmail, R
The most amazing...
...thing I've worked on is prototyping an insurance recommender system for an Italian bank to improve its insurance marketing strategy.
Work Experience
Researcher
TV2.no
- Prototyped a related news article recommender system that achieved a Recall@5 score of more than 50% as evaluated from the perspective of journalists and published the results at the RecSys 2024 workshop conference.
- Used the machine learning ecosystem in Python to accomplish this prototype.
- Involved in building an AI tool for journalists to find related articles quickly and publish them on the news website.
- Employed state-of-the-art large language models to accomplish this task.
Data Analyst
Gruber Logistics
- Built, using an ML stack, an automated fuel event detection system from sensor data generated from hundreds of trucks and deployed it on the Spark cluster to speed up the management of fueling activities of the trucks.
- Contributed to the data product that ran daily, detected all the fueling events of hundreds of trucks, and stored the results in the MongoDB database. The app gave the reporting team a heads-up about the number of payments to be expected from the fuel suppliers.
- Worked on this data app that also gave potential measurements about the fueling activities of the trucks to detect if there was potential fueling fraud happening.
Python Software Developer
Datatellers
- Translated complex credit scoring model, which measured loan worthiness of customers, into Python code.
- Deployed it as AWS Lambda function and exposed it as a REST API endpoint, using AWS API gateway, to be consumed by the front-end team to display customer ratings.
- Worked with modular and efficient code in the speed of NumPy and the size of 250MB of the layer limitation.
Consultant Data Scientist
Sparkasse | Cassa di Risparmio di Bolzano
- Prototyped an insurance recommender system using machine learning classifiers that accurately and transparently ranked customers to be targeted with property, disease, accident, and legal insurance.
- Trained several supervised machine learning classifiers using the Tree-based Pipeline Optimization framework and selected the best explainable model using the cross-validation technique, where the AUC score was the optimized metric.
- Transferred machine learning methodology to the Sparkasse team by guiding them on essential steps required to build ML models, from data preprocessing to feature engineering, feature selection, model building, and finally, model evaluation.
- Produced a detailed research report at the end of the project to communicate the critical findings at each level of the model-building process.
- Deployed the model building and inference pipelines in the Konstanz Information Miner framework for transparency and a better understanding of the Sparkasse team.
Data Analyst
ibex
- Built a multiclass ML model for AT&T and Frontier campaigns that detected customer issues from the transcribed calls with more than 75% accuracy, providing insights into the customer's main pain points.
- Used Python's NLTK package and Stanford's NLP framework to identify customer sentiments about AT&T, Frontier, and Lyft clients and tracked them weekly in the form of Excel reports.
- Managed exploratory data analysis to recommend new benchmarks for customer satisfaction and handling time for new agents to help them become more proficient without too much pressure.
Experience
COVID-19 X-ray Classifier
https://github.com/bilalmahmood1/Covid-DetectorInsurance Recommender System for a Commercial Bank
RESPONSIBILITIES
• Trained several supervised machine learning classifiers using the Tree-based Pipeline Optimization framework and selected the best explainable model using the cross-validation technique, where the AUC score was the optimized metric.
• Transferred machine learning methodology to the Sparkasse team by guiding them on the essential steps required to build ML models, from data preprocessing to feature engineering, feature selection, model building, and model evaluation.
• Produced a detailed research report at the end of the project to communicate the critical findings at each level of the model-building process.
• Deployed the model building and inference pipelines in the Konstanz Information Miner framework for transparency and a better understanding of the Sparkasse team.
Loan Credit Rating for SMEs
I also orchestrated the rating calculation using AWS Step Functions. When customers upload their financial documents into the S3 bucket, that event triggers the parsing of the files. It then prepares the data consumed by the rating calculation endpoint, which stores the results in the DynamoDB database.
URL: https://fdg.mcc.it/rating
Education
Ph.D. in Artificial Intelligence
University of Bergen - Bergen, Norway
Master's Degree in Computer Science
Free University of Bozen-Bolzano - Bolzano, Italy
Bachelor's Degree in Electrical Engineering
Lahore University of Management Sciences - Lahore, Pakistan
Certifications
Deep Learning Specialization
Coursera
Data Analyst Nanodegree
Udacity
Skills
Libraries/APIs
Pandas, XGBoost, OpenAI Assistants API, NumPy
Tools
AutoML, Slack Development, Business Intelligence Development, ChatGPT
Languages
Python, SQL, R
Platforms
Visual Studio Development, Jupyter Notebook, Ubuntu, Data Science, AWS, KNIME, Azure Design
Storage
Database, PostgreSQL, NoSQL
Other
Machine Learning, NLP, Artificial Intelligence, Data Science, Data Science, Supervised Learning, LLM, Deep Learning, Data Visualization, Data Analysis, Statistics, Supervised Machine Learning, Image Recognition, Modeling, Version Control Systems, Communication Coaching, Models, Predictive Modeling, Generative Pre-trained Transformers (GPT), Classifier Development, Teamwork, Data Modeling, Mathematics, Engineering, Recommendation Systems, Gmail, OpenAI GPT-3 API, Chatbot Development, GPT-3, PDF Scraping, Web Scraping, Computer Vision, ChatGPT Prompts
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