Data Science with 1+ years in Data Science & Machine Learning.
AI is analyzing your overall score…
Identifying your key strengths…
Evaluating your skill match against the job requirements…
Assessing your cultural and operational fit
Data Scientist with a background in Statistics and Programming and practical experience in retail fintech and AI-driven diagnostics. Proficient in Python, SQL, PowerBI, and Google Cloud, with strengths in building automated data pipelines and high-accuracy predictive models. Experienced in merchant behavior analysis and in developing AI agents that simplify complex risk data. Skilled at producing investigative reports that translate technical insights into clear, actionable strategies for stakeholders.
Kahuhia Girls High School
Kenya Certificate of Secondary Education (KCSE)
N/A – May 31, 2020
Zindua School
Data Science Core Program · Data Analysis, Visualization and Machine Learning
N/A – June 30, 2025
Kenyatta University
Bachelor of Science · Statistics and Programming
N/A – Present
Duka Technologies
Data Analyst intern
January 1, 2024 – December 31, 2025
India
AI-Powered Gym Member Retention System
June 24, 2026 – Present
Developed an end-to-end diagnostic system to investigate gym member churn and automate engagement strategies using predictive analytics and Natural Language Processing. Built a system to solve the "Retention Gap" by combining behavioral data (how members act) with sentiment data (what members say) to provide a 360-degree view of churn risk. Tuned a Logistic Regression model using demographic and usage data (attendance frequency, membership duration, and payment methods). Achieved a 0.97 accuracy score, demonstrating high reliability in identifying members at risk of discontinuing their memberships. Implemented an NLP Pipeline to analyze member feedback for Sentiment and Topic Identification (Value & Quality, Integration Issues, and High Engagement signals). Designed a rule-based logic to turn these text-based insights into actionable recommendations for gym management. Engineered a Streamlit Dashboard that serves as the central hub for monitoring retention metrics, visualizing churn probabilities, and reviewing feedback diagnostics in real-time. Stored models as .pkl files for efficient deployment and integrated Pandas and NumPy for automated data preprocessing.
View ProjectMaven Fuzzy Factory E-Commerce Optimization Project
June 24, 2026 – Present
Led the strategic analysis to maximize Revenue per Session (RPS) for an online toy retailer by diagnosing conversion funnels and identifying high-impact growth levers. Addressed "value efficiency" constraints where strong traffic failed to scale due to critical checkout bottlenecks, low average order values (AOV), and inefficient marketing spend. Analyzed the full conversion funnel to identify a "Catastrophic Leak" at the billing stage, where a 0.0% completion rate resulted in an immediate $216,000 annual revenue loss. Quantified drop-offs at each funnel stage (Sessions to Order Confirmation) using Pandas to prioritize high-impact engineering and UX interventions over broad traffic acquisition. Formulated and executed a one-tailed T-test (with Welch's correction) to compare Average Order Value (AOV) between new and returning customers. Statistically validated (p-value = 0.0044) that repeat customers generate a higher AOV, providing the data-driven justification for a transition from acquisition-heavy to retention-focused strategies. Engineered a "Four-Pillar Roadmap" projected to unlock $1.65 Million in annual revenue by driving structural RPS improvement from $4.10 to $4.39. Designed a channel reallocation strategy to shift spend from low-performing sources ($0.71 RPS) to high-value traffic, increasing efficiency without increasing the total marketing budget. Quantified the "AOV Trap," identifying that 76% of orders were single-item and proposing a bundling strategy to capture the $38.43 revenue premium found in multi-item orders.
KenyaRE AI Hackathon Project 2025
January 1, 2025 – December 31, 2025
Led a team in Developing an AI-powered risk assessment platform to automate the collection and analysis of fragmented insurance data for the Kenyan reinsurance market. Addressed the manual bottleneck in reinsurance by aggregating siloed data from the KNBS, IRA, CBK, and Ministry of Transport. Engineered Selenium web scrapers in Python to continuously collect risk data from government portals, eliminating the need for manual data requests. Leveraged AWS S3 for secure storage and management of large-scale risk datasets and model artifacts. Implemented a RAG (Retrieval-Augmented Generation) system using LangChain, Faiss, and Google Gemma 3B to build a "Virtual Reinsurance Analyst" chatbot. Developed regression and clustering models in TensorFlow to segment risks and predict county-level exposure. Built an interactive Streamlit dashboard featuring county-level heatmaps to visualize risk distribution (Aviation, Transport, Mortgage Default, etc.). Designed "hover tooltips" and risk score tables to give underwriters an instant view of exposure across Kenya's 47 counties.
View ProjectGenerative AI & AI Agent Development
Unstacked Labs
June 1, 2026 – Present
Data Science Core Program
Zindua School
January 1, 2025 – Present
Kenya Certificate of Secondary Education (KCSE)
Kahuhia Girls High School
January 1, 2020 – Present
Cultural Fit Analysis
The candidate's project diversity, ranging from gym member retention to reinsurance risk assessment and e-commerce optimization, showcases a broad interest and adaptability to different domains. Their academic and professional projects align well with a Data Science role, demonstrating a clear career path and passion for the field. The use of Agile methodologies in their internship and project work indicates an understanding of modern development practices and a collaborative approach, which are positive indicators for cultural fit within a dynamic team.
Soft Skills & Operational Fit
The candidate demonstrates strong problem-solving abilities, evidenced by their detailed project descriptions addressing specific business challenges. Their experience in leading teams (KenyaRE Hackathon) and collaborating with product/finance teams (Duka Technologies) indicates good teamwork and communication skills. The ability to manage real-time data requests and deliver simultaneous reports suggests adaptability and efficiency in high-tempo environments. The candidate's involvement in community engagement (Civic Signal Hackathon facilitator) also points to a proactive and collaborative mindset.