Data Engineer with 3.5 years of experience building production AWS data platforms and GenAI solutions. Built a RAG-based analytics system that took call-quality evaluation at flydubai from 2–3% manual sampling to 100% au…
Data Engineer with 3.5 years of experience building production AWS data platforms and GenAI solutions. Built a RAG-based analytics system that took call-quality evaluation at flydubai from 2–3% manual sampling to 100% automated coverage and recognized with the flydubai Business Enabler award. Demonstrated expertise in cloud-based data architectures, machine learning, advanced analytics, and enterprise data warehousing. Strong in Spark on EMR, Kafka (MSK), Redshift, Hive and serverless orchestration, with a machine learning background (MSc, UCL) and a track record of translating business requirements into governed, scalable pipelines.
<p>Designed and implemented production-grade data platforms and ETL pipelines on AWS, supporting large-scale ingestion, transformation, and analytics across multiple business domains.<br>
Developed a RAG-based GenAI analytics solution using Amazon Bedrock, grounding Genesys interaction assessments in evaluation guidelines and compliance policies; implemented structured prompting and tool/function calling to automate scoring across 12+ parameters.<br>
Received the flydubai Business Enabler award for the solution’s innovation and impact, increasing interaction-evaluation coverage from approximately 2–3% manual sampling to 100% automated analysis.<br>
Developed scalable Spark-based processing frameworks using EMR / EMR Serverless to transform data from APIs, streaming sources (MSK Kafka), and batch feeds into governed Hive tables and Redshift analytics models.<br>
Created and optimized Redshift schemas and Hive-based data lakes, enabling high-performance analytics, BI reporting, and downstream machine learning workloads through incremental loads and partitioning strategies.<br>
Automated end-to-end data workflows using AWS Lambda, Step Functions, DynamoDB, S3, and Glue, enabling configuration-driven pipelines with improved reliability and reduced operational overhead.</p>
<p>Spearheaded the development of a comprehensive data augmentation pipeline using deep learning models in PyTorch for complex time series dataset, resulting in a 20% improvement in anomaly detection model accuracy.<br>
Conducted extensive research and analyzed survey papers to identify optimal models and provided detailed reports to concerned stakeholders comprising the findings.<br>
Utilized Python libraries (numPy, scikit-learn etc) to perform in-depth exploratory data analysis and preprocessing for data modeling on large datasets, ensuring data quality and reliability for subsequent anomaly detection.<br>
Designed and executed a rigorous project plan, including meticulous experimentation and clear PowerPoint presentations to communicate complex methodologies to non-technical stakeholders.<br>
Successfully deployed data augmentation approaches (jittering, scaling, permutation) and deep learning models, Autoencoders - LSTM & Variational, and Wasserstein Generative Adversarial with gradient penalty) for time series data.<br>
Led subsequent performance evaluations such as accuracy, F1 score, recall and precision against alternative models to quantify efficacy.<br>
Applied various machine learning algorithms for anomaly detection, including isolation forests and one-class SVM, followed by comprehensive performance evaluation to quantify efficacy.</p>
<p>Worked with Product Information Management (PIM) team to develop components for Schneider Electric's global master data management system using JavaScript.<br>
Facilitated communications with cross-cultural teams comprising of product owners and QA testing team for gathering business requirements and scope, crafting user stories, and devising solutions to translate them into actionable insights.<br>
Actively participated in code review processes to maintain code quality, share knowledge, and adhere to industry standards (BMECat, ETIM etc).<br>
Utilized version control (Git) to track code changes, ensure codebase stability, and facilitate collaboration with the PIM team.<br>
Participated in agile methodologies (scrum), driving on-time project completion and fostering team collaboration through backlog management, sprint planning and prioritization of user stories and EPICS in Jira.<br>
Developed Tableau dashboards to track KPIs and monitor system performance, identifying bottlenecks and inefficiencies that led to a 39% reduction in system breakdowns.<br>
Proficiently employed MySQL and Microsoft Excel for data migration tasks, meticulously verifying the accuracy and consistency of transferred data to ensure seamless transitions.</p>
The AI-first Recruitment Operating System that unifies and automates the entire hiring lifecycle for modern enterprises worldwide.
Dr Job FZ LLC. 2026 © All Rights Reserved
