ABID Koceila

𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 & 𝐑𝐞𝐜𝐨𝐦𝐦𝐞𝐧𝐝𝐚𝐭𝐢𝐨𝐧𝐬 Designed and deployed advanced customer segmentation and targeting strategies. Built recommender systems leveraging collaborative filtering, content-based, and hybrid models, delivering personalized product suggestions that drove measurable increases in user engagement. 𝐆𝐞𝐧𝐞𝐫𝐚𝐭𝐢𝐯𝐞 𝐀𝐈 & 𝐍𝐋𝐏 Integrated Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) into production systems to power personalized product recommendations, automated categorization, and marketing content generation. Deployed LLM capabilities across both real-time (AWS Lambda) and batch workflows, leveraging tools such as OpenAI, LangChain, LangSmith, and vector stores to deliver scalable, high-performance solutions that significantly enhanced customer experience and operational efficiency. 𝐃𝐚𝐭𝐚 𝐀𝐧𝐚𝐥𝐲𝐭𝐢𝐜𝐬 Conducted data analytics for email marketing campaigns, developing interactive dashboards to track KPIs such as click-through rates, conversion rates, and ROI. Implemented A/B testing strategies to optimize campaign performance, driving measurable improvements in user engagement and revenue. 𝐁𝐢𝐠 𝐃𝐚𝐭𝐚 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 Architected, optimized, and maintained large-scale data processing workflows on AWS and Databricks using python and spark, ensuring efficient ETL pipelines, high data quality, and reduced processing time for analytics and machine learning workloads.

𝐅𝐞𝐚𝐭𝐮𝐫𝐞 𝐄𝐧𝐠𝐢𝐧𝐞𝐞𝐫𝐢𝐧𝐠 Applied advanced statistical and signal-processing techniques to extract meaningful indicators from industrial vibration data, which significantly enhanced condition monitoring and predictive maintenance capabilities. 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 & 𝐃𝐞𝐞𝐩 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 Developed an innovative fault identification method using anomaly detection algorithms like One-Class SVM (OCSVM) and Autoencoders (AE), combined with Generalized Linear Models (GLM) for precise failure prediction. Employed ensemble deep learning models combining Convolutional Neural Networks (CNN) and Long Short-Term Memory (LSTM) architectures, achieving improved evaluation metrics over previous approaches. 𝐏𝐫𝐨𝐠𝐫𝐚𝐦𝐦𝐢𝐧𝐠 Proficient in Python programming with expertise in libraries and frameworks such as pandas, scikit-learn, TensorFlow, and Keras for machine learning and deep learning development. 𝐒𝐜𝐢𝐞𝐧𝐭𝐢𝐟𝐢𝐜 𝐂𝐨𝐦𝐦𝐮𝐧𝐢𝐜𝐚𝐭𝐢𝐨𝐧 Authored and presented scientific papers at leading international conferences, including ICML, ICMLA, and the PHM Society, contributing to advancements in AI and predictive maintenance research.

𝐃𝐚𝐭𝐚 𝐂𝐨𝐥𝐥𝐞𝐜𝐭𝐢𝐨𝐧 Gathered sensor data from smart home environments equipped with GSM-enabled devices. These devices were deployed in the homes of elderly individuals to monitor daily activities. 𝐃𝐚𝐭𝐚 𝐏𝐫𝐨𝐜𝐞𝐬𝐬𝐢𝐧𝐠 Cleaned and preprocessed smart home sensor data to accurately recognize and classify activities of daily living (ADL). 𝐌𝐚𝐜𝐡𝐢𝐧𝐞 𝐋𝐞𝐚𝐫𝐧𝐢𝐧𝐠 Developed activity recognition models using Hidden Markov Models (HMM) to detect abnormal behaviors in elderly individuals, which enhanced early intervention capabilities.