Fedi Wannes

Project Name: LLM Integration and Optimization for Big Data ● Led testing and optimization of local language models (LLM). ● Collaborated in a team to ensure high-quality code and prompt engineering. ● Developed a chatbot using local models and Hugging Face API. ● Implemented SQL agents for natural language database interaction. ● Automated performance reporting and integrated AI technologies within Cloudera, Hive, and PostgreSQL environments. Main Technologies: Python, LangChain, Hugging Face API, Cloudera, Hive, PostgreSQL, PgVector, Faiss, LLama2, Pinecone Applied Method: Agile Scrum

- Led the migration of on-premise Cloudera data workflows to the Databricks platform. - Collaborated on planning, designing, and executing the migration, including adapting Java Spark code to Delta. - Tested JAR files, managed ingestion processes, and orchestrated jobs in Databricks workflows. - Configured schedulers, triggers, and alerts, and redesigned dashboards from Splunk to Databricks. - Worked on Unity Catalog for organizing schemas and data flows. - Ensured alignment with team needs and adhered to recommendations from Databricks engineers.

Project Name: SEAMCAT Open-Source ● Contributed to the development of new features for a Monte-Carlo simulation software ● Improved the efficiency of statistical modelling for various radio interference scenarios ● Mentored junior developers and ensured code quality on the project ● Reviewed code and debugged errors to improve performance. Main Technologies: Core Java, Java Swing, Java Hibernate,Apache Maven, Sqlite Applied Method: Agile Scrum

Project Name: Risk Calculation Application and Data Processing Optimization ● Led pull request validation and managed Git branching for the risk calculation application. ● Collaborated with clients to validate solutions and integrate event triggering via Kafka, enhancing real-time data processing. ● Utilized Spark Launcher via Java, optimizing large-scale data processing tasks to schedule and trigger the job runs. ● Replaced the existing Spring Batch application with Spark-based solution, significantly reducing calculation time and improving performance. ● Configured and managed Spark clusters, ensuring seamless integration with the data lake and enabling efficient data validation processes. Main Technologies: Apache Spark, Java, Spring Data, Kafka, Spark Launcher, Spring Batch, Data Lake, Git Applied Methods: Scrum, Kanban

Development of an internal software in Altran which makes a synchronization of Jira tickets from Clients Jira under the Local Jira

Project Name: Wescore football live score & news [In-Car app] ●Participated in the project from its inception, including the design of the data model ● Implemented an XML reader for real-time data sources ● Regularly updated the database on Amazon S3 ● Developed the integration between the backend and frontend, ensuring seamless communication of results ● Utilized Spring Security for JSON Web Token (JWT) implementation and optimized the system after penetration test

Taking an active role in the pursuit of own goals. Having a voice in the local communities. Development of communication skills. Importance of teamwork. Adaptation to a new culture. Presentation skills. Facilitating and training experience.

Data analyst operating on databases using Python programming language (Pandas, Numpy, MAtplotlib) libraries expert

Integration of business processes in web application models using BonitaSoft open source platform | BPM