AI/ML Engineer with 4+ years of experience designing and deploying machine learning and LLM-driven systems in production environments. Specialized in Retrieval-Augmented Generation (RAG), NLP pipelines, and low-latency A…
AI/ML Engineer with 4+ years of experience designing and deploying machine learning and LLM-driven systems in production environments. Specialized in Retrieval-Augmented Generation (RAG), NLP pipelines, and low-latency API services using Python, PyTorch, and FastAPI. Proven ability to improve retrieval accuracy, optimize inference performance, and deliver scalable AI solutions handling high-volume enterprise workloads.
Optimized system performance by balancing latency vs retrieval accuracy trade-offs, improving overall response time without degrading output quality., Containerized services using Docker and CI/CD pipelines, reducing deployment issues by 40%., Enforced RBAC and secure API access for sensitive data, ensuring compliance with enterprise data governance standards., Utilized AWS S3 and EC2 to support scalable data processing and distributed workloads.
Designed a RAG architecture combining FAISS indexing and hybrid retrieval, improving document retrieval accuracy by 32% across policy and claims datasets (~2M+ records).
Built embedding and chunking pipelines aligned with document structure, improving response relevance and reducing context loss in LLM outputs.
Implemented hybrid retrieval (vector + keyword) with reranking, reducing unsupported responses by 28% in compliance-driven workflows.
Developed reusable prompt design patterns and context injection strategies to standardize outputs across underwriting and claims use cases.
Deployed NLP and computer vision models via FastAPI services, supporting 5K–7K daily requests with
Designed a RAG architecture combining FAISS indexing and hybrid retrieval, improving document retrieval accuracy by 32% across policy and claims datasets (~2M+ records)., Built embedding and chunking pipelines aligned with document structure, improving response relevance and reducing context loss in LLM outputs., Implemented hybrid retrieval (vector + keyword) with reranking, reducing unsupported responses by 28% in compliance-driven workflows., Developed reusable prompt design patterns and context injection strategies to standardize outputs across underwriting and claims use cases., Deployed NLP and computer vision models via FastAPI services, supporting 5K–7K daily requests with
Developed churn prediction and customer segmentation models using XGBoost and Scikit-learn on large-scale datasets, improving retention by 20%.
Engineered feature pipelines including behavioral and temporal features, improving model performance by 22%.
Built PySpark-based ETL pipelines for structured and semi-structured data, reducing data processing time by 30%.
Implemented NLP pipelines using BERT and spaCy for sentiment analysis and entity extraction, improving classification accuracy by 18%.
Introduced data validation and schema enforcement frameworks, reducing data inconsistencies by 20%.
Deployed models through Flask APIs for integration with business systems and analytics platforms.
Automated reporting workflows using Power BI and SQL, reducing reporting turnaround time by 35%.
Optimized database queries and data workflows using PostgreSQL and MongoDB.
Built classification and regression models using Scikit-learn, improving prediction accuracy by 15% across structured datasets.
Performed data preprocessing and feature engineering, reducing data quality issues by 25% and improving model input reliability.
Developed REST APIs using Flask to enable real-time inference and system integration.
Optimized SQL queries and data extraction workflows, improving data retrieval speed by 30%.
Created Tableau dashboards for model insights and KPIs, reducing manual reporting effort by 25%.
Collaborated with cross-functional teams to refine features and evaluation strategies, improving model consistency by 18%.
Optimized SQL queries and data extraction workflows, improving data retrieval speed by 30%., Created Tableau dashboards for model insights and KPIs, reducing manual reporting effort by 25%., Collaborated with cross-functional teams to refine features and evaluation strategies, improving model consistency by 18%.
نظام التوظيف التشغيلي المعتمد على الذكاء الاصطناعي أولاً، الذي يوحّد ويؤتمت دورة التوظيف بالكامل للمؤسسات الحديثة حول العالم.
د.جوب منطقة حرة ذ.م.م. 2026 © جميع الحقوق محفوظة
