Malek Slokom

- Automated central reporting API configuration by leveraging warehouse metadata and advanced DBT modeling, reducing configuration errors by 80% and accelerating time-to-market for new reporting use cases. - Designed and led the integration of DBT macros into ClickHouse models, standardizing transformation logic across large-scale real-time datasets with over 5,000 daily queries. - Built and maintained observability dashboards in Grafana to monitor pipeline performance and data quality across critical components of the platform. - Collaborated with product owners in an Agile feature team to deliver data platform enhancements aligned with client reporting needs.

- Developed a web-based solution to automate and optimize computational task execution for researchers. - Integrated Kubernetes to manage and balance workloads across powerful infrastructure resources. - Designed a user-friendly interface for task submission, execution tracking, and real-time resource monitoring. - Implemented a project-based resource allocation system to ensure fair usage and prioritization among researchers. - Collaborated closely with researchers to understand their needs and tailored the solution accordingly.

- Designed and implemented a machine learning model using supervised learning algorithms to predict fiscal indiscipline risks, improving accuracy in forecasting potential issues. - Created a "human-in-the-loop" web application to streamline the integration of the machine learning model into decision-making processes, allowing for real-time user input and feedback to refine model predictions.

Worked as a freelance developer on the Karine.so and Arownd.me projects.

- Implemented a comprehensive data warehouse to facilitate advanced crime analysis and prediction, leveraging my talent in data architecture and management. - Conducted extensive data preprocessing to clean, transform, and organize raw data, ensuring high-quality inputs for subsequent analysis. - Executed feature engineering techniques to identify and extract relevant features from the dataset, enhancing the model's predictive capabilities and overall performance. - Performed hyperparameter tuning to optimize machine learning models, improving accuracy and reliability in predicting crime patterns and trends. - Developed a user-friendly mobile and web application tailored for police officers, enabling real-time access to crime data and analytical insights to support proactive policing efforts.