Ahmed Ewis

Developed a custom chatbot using OpenAI's LLM and LangChain, achieving a 70% reduction in manual query handling and integration with Microsoft Teams.
Utilized Docker and GitHub Actions for efficient, seamless production deployments.
Oversaw Azure deployments, addressing cloud setup and infrastructure challenges.
Developed a sentiment analysis model with deep learning, targeting nuances in the Kuwaiti dialect.
Resolved 10+ business cases, providing the marketing team with insights that enhanced campaign effectiveness by 20% based on detailed statistical analysis.
Headed a data scientist team, driving a 25% rise in both customer satisfaction and operational efficiency.

1. Directs a team of Data Scientists to design, develop, and implement end-to-end machine learning production pipelines (data exploration, sampling, training data generation, feature engineering, model building, and performance evaluation)
1. Ensures that data pipelines are scalable, repeatable, and secure, and can serve multiple users within the company
2. Identify, oversee and deploy advanced analytics use cases and support integration with front-end digital tools
3. Enables big data and batch/real-time analytical solutions that leverage emerging technologies
4. Collects, parses, manages, analyzes and visualizes large sets of data using multiple platforms
5. Translates complex functional and technical requirements into detailed architecture, design, and high performing software
6. Codes, tests, and documents new or modified data systems to create robust and scalable applications for data analytics
7. Implements security and recovery tools and techniques as required
8. Works with Data Scientists and developers to make sure that all data solutions are consistent
9. Ensures all automated processes preserve data integrity by managing the alignment of data availability and integration processes

Developed a range of machine learning models encompassing credit card and personal finance propensity models, customer churn prediction, budget forecasting, product recommendation systems, and customer segmentation analysis.
Enhanced conversion rates by 30% using A/B testing, leading to a 10% profitability boost and refined customer conversion assessments.
Introduced and institutionalized a novel customer segmentation method via clustering models across the bank.
Utilized advanced tools for data analysis, boosting processing efficiency by 30%.

Demonstrated expertise in Generative Adversarial Networks (GANs) using PyTorch, supported by various deep learning techniques.
Utilized leading pre-trained models of DALLE, BERT, and CLIP for diverse AI applications.
Fine-tuned models on custom datasets, enhancing task performance by 25%.
Advanced speech recognition using NLP and Microsoft & Google APIs, achieving 20% better accuracy.
Generated custom NFTs with the stable diffusion model, deploying via AWS SageMaker.

Maintained all technology documentation, providing updates and addendums when required.
Developed new compounds and formulas based on strict customer requirements.
Leveraged data analysis tools to assess historical data, employing descriptive and diagnostic analyses to forecast close to 95% of the optimal pricing and conditions for multiple projects.
Crafted an Excel costing template streamlining price estimation for employees, resulting in significant time savings and precise calculations of up to 90%.
Developed processes based on critical needs, preset efficiency standards, and cost reduction measures up to 50%.

Actively participated in Kaggle competitions—affiliated with Google LLC—contributing to the global data science community.
Achieved Top 15% in OSIC Pulmonary Fibrosis (300/2097) using CNN to predict lung function decline.
Ranked in the Top 19% for Cassava Leaf Disease Classification (732/3900) using CNN for disease detection on leaf images.
Placed in the Top 24% in BirdCLEF 2021 challenge (189/816) for bird call identification using deep learning with CNN.
Secured Top 20% in Rainforest-Connection-Species-Audio-Detection (218/1143) leveraging CNN techniques.

Helped students to recognize learning issues evidenced in one-to-one support instruction.
Data Cleaning and data mining techniques to assist the research students.
Increased the accuracy of models by 40% and helped students gain a higher understanding of the research by providing 60% more research support.
Streamlined research process and reduced data processing time by 70%.
Improved data collection methods, resulting in a 50% increase in accuracy.

Undertook a virtual internship in Data Analytics Consulting, focusing on data analysis and reporting.
Enhanced data analysis accuracy by 35% using Data Quality Assessment methods, achieving 80% accuracy in visual impairment estimations.
Presented Data Insights reports with key findings derived from descriptive and inferential statistics to the management.
Developed a dashboard aligned with KPMG's standards, reflecting 3,490 customers and achieving a 10.93M KPI.

Gained a thorough understanding of data science, encompassing data mining, statistical analysis, and machine learning.
Acquired proficiency in tools of Python, R, SQL, PowerBI, Tableau, and Git for the entire data science workflow.
Trained on a breadth of data science packages, from data entry to deep learning and AI.
Completed 28 intensive courses, each reinforced with exercises and projects.
Deployed predictive models in data pipelines, boosting accuracy by 20% through feature engineering, tuning, and ensemble models.