Lead AI/ML engineer

- Achievements

• Architected and built a complete MLOps platform from requirements gathering through production deployment: conducted stakeholder interviews across data science, ML, and data engineering teams; designed governance structure with RACI matrix; led technical design and implementation of a scalable, cloud-native ML infrastructure on AWS.
• Developed a proprietary ML automation library and framework enabling data scientists to build, train, and deploy models with minimal infrastructure knowledge. The library provides end-to-end tooling for model development through production: SageMaker training job orchestration with automated hyperparameter tuning, model registry and versioning, integration with Snowflake/dbt feature pipelines for automated feature fetching, batch and real-time inference abstractions, automated score writebacks to Snowflake, and model performance monitoring. Fully integrated with GitLab CI/CD to standardize training, validation, and deployment workflows. Dramatically reduces time-to-productionization and allows data scientists to focus on model development rather than infrastructure.
• Designed and built an enterprise-scale feature store serving both big data (high-volume batch features) and small data (real-time, low-latency features) covering the full spectrum of data generated by the data team: product holdings, consumption metrics, video watching behavior, and cross-domain datasets. Established naming conventions, dbt project structure, CI/CD validation pipelines, and Dagster scheduling (with differentiated schedules for batch vs. real-time data) to enable reliable, versioned feature materialization. Created comprehensive documentation and self-service contribution guidelines so data scientists can independently add, modify, and maintain features. Provides reusable, tested features across all ML projects and enforces consistent data definitions and quality standards.
• Established production ML operations through Dagster-based orchestration, automated model monitoring dashboards (Streamlit), performance tracking, data quality alerts, and incident response workflows. Reduced operational friction and model training time by standardizing deployment patterns and automating routine tasks across the ML lifecycle.
• Created an ML project template and onboarding framework with best-practice structure including Docker containerization, GitLab CI/CD pipelines, dependency management via uv, folder conventions, and testing standards. Enables new projects to go from experimentation to production in days rather than weeks, dramatically reducing time-to-value.
• Built large-scale data processing pipelines using PySpark on AWS EMR for feature preparation, batch scoring, and continuous retraining workflows. Handles millions of records and complex transformations across multiple data sources, enabling scalable, repeatable data pipelines.
• Built a FastAPI-based LLM assistant integrated into Microsoft Teams for data scientists to query feature metadata, data-quality status, and compliance constraints in real-time. Uses multi-step LLM orchestration and prompt engineering to ground responses in live feature store metadata, reducing manual inquiries and enabling self-service governance.
• Designed and implemented an end-to-end automated GDPR and Right to Be Forgotten (RTBF) compliance system that identifies and removes customer data across distributed ML datasets, feature stores, and model artifacts. Coordinated with legal, data protection, and data engineering teams to define deletion workflows; built automated pipeline to execute purges on schedule; and created audit logging for regulatory compliance. Eliminates manual, error-prone processes and ensures data privacy standards are met at scale.
• Led migration from legacy ML processes toward scalable cloud infrastructure. Championed adoption of modern tooling, documentation standards, and collaborative practices across distributed ML teams.

Data Science and Software Engineering Researcher

- Achievements

• Conducted large-scale empirical research on software development practices by analyzing 188,000+ pull requests and millions of commits across hundreds of GitHub projects. Built data collection pipelines using GitHub APIs and CouchDB; implemented comprehensive data cleaning and validation workflows.
• Developed machine learning models for software engineering tasks, including bot detection classifiers using scikit-learn and Keras, software quality prediction models, and behavioral analysis systems. Created ground-truth datasets and validated results through rigorous cross-validation and statistical testing.
• Applied advanced NLP and data science techniques using Pandas, Seaborn, Scipy, NLTK, and spaCy to extract insights from GitHub data and software repositories.
• Processed large-scale data using PySpark and Microsoft Azure infrastructure for feature engineering, aggregation, and batch analysis.
• Created reproducible research artifacts including Jupyter notebooks, replication packages, cleaned datasets, and trained models published for community use.
• Published 11+ peer-reviewed papers with 547+ citations in top-tier venues (IEEE Software, Journal of Systems and Software, SANER, ICSME, FSE). Secured 4 years of research funding through FNRS-FWO Excellence of Science program.

Software Developer

- Achievements

• Developed and maintained multiple enterprise applications using Java Spring ecosystem (Spring MVC, Spring Boot, Spring Security, Hibernate, JPA) and C# (WPF, MVVM, Prism), delivering robust, scalable systems for diverse business domains.
• Designed and implemented biometric applications including authentication systems and identity verification workflows; translated complex security requirements into production-grade solutions.
• Architected a large-scale distributed front-end application using C#, WPF, and MVVM patterns with SOAP services integration, handling high-volume concurrent users and complex data interactions.
• Developed enterprise REST web services providing APIs for system integration and data access across multiple applications; ensured high performance, reliability, and maintainability.
• Established software engineering best practices including comprehensive unit and integration testing, automated CI/CD pipelines with Docker containerization, code coverage analysis, and peer code review processes. Created documentation standards and mentored team members on quality practices.
• Led technical design and architecture discussions across projects, participating in system design, technology selection, and implementation planning with stakeholders and cross-functional teams.

Software Developer

- Achievements

• Designed and developed 2 Android applications (JAVA - SQLite - rest web services)
• Designed and developed more than 10 websites for different companies.
• Designed and developed application with most recent design patterns
• Followed scrum methodology in software development process

IT Instructor (part time)

Fundamental and advanced programming courses