*Managing a team of data scientists & subject matter experts, deploying AI solutions, analytical dashboards and reports for executive management, facilitating tracking of KPIs and transformation projects.*Project Management of multiple AI Based, decentralized E2E Proactive solution initiatives and POC Models. Ensuring a timely delivery and tangible business values.*Conducting stakeholder Requirement Gathering, Performing Business Analysis finalized into Technical Requirements Documents (TRD ) & Business Requirements Document (BRD) to be readily used in Development Cycle*Managing data science team, for gathering and performing end-to-end data analysis, featuring engineering, manipulation and model selection, ensuring completion within Time-to-Market key performance indicators (TTM KPI) with superior post deployment accuracy.*Engineered a detection and monitoring algorithm utilizing a blend of pattern recognition and classification models, detecting anomalies across a multi digital platform automated solutions.*Developed a statistical/ML model detecting improper use of company platforms, primarily related to employee behavior, prompting revisions to existing KPIs and the introduction of new KPIs across various departments.Key Achievement:*Applied deep learning (LSTM, CNN) and statistical models to identify abnormal patterns in technical consumer complaints with over 2500 features, maintaining an accuracy of 90% or higher, leading to an 80% decrease in root cause analysis (RCA) delivery time.*Designed classification algorithms to detect outages in various regions across the UAE, achieving an 85% reduction in detection time, leading to a reduction of OPEX and CAPEX costs.*Developed and Deployed over a dozen machine learning models across multiple channels.*Developed and deployed a predictive model for hourly customer complaint volumes up to 10 days in advance, resulting in a 20% reduction in complaints within the first three months of deployment.

*Developed intensive mathematical and statistical models to represent complex systems, used as the governing equations for unsupervised/reinforcement learning AI/ML solutions, focusing on power systems, control systems, and renewable energy. These models are characterized by interdependent equations, where each element influences other components within the sandboxed simulations.*Established mathematical frameworks serving as the foundation for optimization models' fitness functions and loss calculations.*Employed evolutionary algorithms, including Genetic algorithms, particle swarm optimization, and Ant colony optimization techniques, for single-objective optimization tasks. Objectives include emission reduction, power loss minimization, voltage drop minimization, profit maximization, and deferring system upgrades to reduce cash outflows, thereby freeing up capital for alternative investments.*Implemented a non-dominant sorting algorithm tailored for multi-objective optimization problems. Utilized crossover and mutation operators suitable for diverse objective functions with varying calculation bases. Challenges include weighting the outputs of each fitness function: The primary goal is to generate a set of non-dominant solutions and identify the optimal solution using Pareto front representation.*Engaged in active research and publication within the energy sector. Focus areas include optimal allocation of resources for cost-benefit analysis and assessing the impact of integrating distributed generation units on existing protection systems.