Hamze
From United Kingdom (UTC+1)
Hamze – Python, Databricks, DBT
Hamze is a strong senior Data Engineer with 9+ years in data and machine learning across automotive manufacturing, insurance, energy, and banking. He builds enterprise data platforms and pipelines on Azure (Data Factory, Data Lake), Databricks, and Palantir Foundry using Python, PySpark, and SQL, and adds ML and LLM solutions on top. He has led data science and engineering teams and is used to turning fragmented data into reporting and models the business can rely on.
Main highlights:
- Cut fraud incidents by 30% with an LLM-based fraud detection system for an insurance provider
- Improved revenue assurance accuracy by 25% by rebuilding financial reporting pipelines to meet FCA compliance
- Cut onboarding time for new projects by 25% by standardising global master data for an automotive manufacturer
9 years of commercial experience in
Main technologies
Additional skills
Direct hire
PossibleReady to get matched with vetted developers fast?
Let’s get started today!Experience Highlights
Principal Data Engineer
A global automotive parts manufacturer (vehicle seating systems) had data scattered across regions, plant systems and IoT sources, which slowed down operational decisions and made launching new projects slow. Hamze led the design of an enterprise data platform on Palantir Foundry that brought these sources into a single Just-In-Time process framework and a global master data catalogue, with governed PySpark pipelines feeding analytics and ML use cases such as predictive maintenance and supply chain analytics. The work reduced process latency, improved operational decision-making accuracy by 20% and cut onboarding time for new projects by 25%.
- Led the design and delivery of enterprise data solutions on Palantir Foundry, integrating IoT, manufacturing and enterprise data
- Built a Just-In-Time process framework from global data sources, improving operational decision-making accuracy by 20%
- Standardised master data across regions, cutting onboarding time for new projects by 25%
- Architected scalable, secure data pipelines and governance frameworks for data quality and compliance
- Set up Foundry workflows for AI/ML model deployment in predictive maintenance, logistics and cost-reduction use cases
- Worked with global cross-functional teams to pick high-impact data use cases
- Mentored junior engineers and set best practices for Foundry development
Data Engineer/Scientist (Energy Trader) with Machine learning specialism
A UK renewable energy supplier needed sharper tariff pricing, tighter credit risk control and data-driven testing of new wind turbine materials. Hamze built LLM-based models that adjust energy tariffs using consumer sentiment and real-time energy data, developed credit risk models and scorecards for residential and business customers, and created ML surrogate models with time series analysis to test new turbine materials against performance benchmarks, working with Python, PySpark, Databricks, Azure Data Factory and AWS. Pricing accuracy improved by 20%, financial defaults fell by 5%, and operational efficiency in one region rose by over 18%.
- Built LLM-based tariff prediction models using consumer sentiment and real-time energy data, improving pricing accuracy by 20%
- Developed credit risk models and scorecards for residential and business customers, cutting financial defaults by 5%
- Created ML surrogate models and time series forecasts to test new wind turbine materials, lifting regional operational efficiency by over 18%
- Optimised energy commodity trading models, including better weather pattern prediction
- Built first- and third-party fraud prevention strategies and managed bad debt provisioning
- Produced credit risk reporting and optimised the commodity database and warehouse sampling
- Ran internal seminars on the ethical use of machine learning
Data Analyst (Machine learning specialist)
A large UK building society (mutual retail bank) was seeing drift in its credit risk decision models, which weakened risk assessments across application, behavioural and collections cycles. Hamze identified and corrected the drift using regression and machine learning techniques in Python and scikit-learn, restoring predictive accuracy and making the models more stable for future financial cycles. The fix reduced the risk of defaults by 5%.
- Detected and corrected drift in credit risk models, reducing default risk by 5%
- Developed, maintained and monitored credit risk decision models across application, behavioural and collections cycles
- Supported affordability solutions across retail lending portfolios
- Built analytical tools to identify economic crime
- Applied regression modelling and machine learning techniques in model development
Data Scientist/Engineer
An online learning company running live sessions on Zoom and Teams wanted hosts to keep audiences more engaged. Hamze built a real-time NLP platform integrated with large language models that helps hosts keep discussions relevant and varied, and optimised the AWS ETL pipelines (Spark, Redshift) behind it for real-time data handling. Audience engagement grew by 30% and pipeline latency dropped by 40%.
- Built a real-time NLP platform with LLM integration for live virtual sessions, increasing audience engagement by 30%
- Optimised AWS ETL pipelines for real-time data, cutting latency by 40%
- Built predictive models with machine learning frameworks such as TensorFlow and PyTorch
- Ran statistical analysis on large datasets to find insights that improve business performance