collins macharia | Data Scientist | Analyst | ML Engineer | Full Stack Developer

collins macharia

Data Scientist | Analyst | ML Engineer | Full Stack Developer

Based in Nairobi

About

I am a results-driven Data Analyst | Scientist | Engineer| Full-stack Developer with expertise in transforming raw data into actionable intelligence and building the infrastructure that makes it possible. With a foundation in financial engineering and advanced training in data science, I specialize in predictive modeling, statistical analysis,data analytics, data visualization and machine learning to solve complex business challenges. More recently, I have expanded into data engineering—designing ETL pipelines, integrating MongoDB with fintech platforms, and automating reporting systems that scale across large client bases—while also leveraging my software development experience from building a customer churn prediction app to leading full-stack development in my most recent senior role.

Skills

Projects

Fraud Detection and Analysis in Banking Transactions

This project focuses on detecting fraudulent banking transactions using machine learning. Using a dataset of 116,201 transactions from Kaggle, I conducted data preprocessing, Exploratory Data Analysis, and anomaly detection. The Isolation Forest algorithm identified 5,565 fraudulent transactions.

Project-Lists-Management-App

This Scalable Project Management Web App is designed to streamline task and project tracking for individuals or small teams across multiple categories such as Home, Work, and Lifestyle. Developed using Flask and PostgreSQL, and deployed on AWS EC2.

Fuel Cost Estimation & Route Optimization API

This project was designed to solve a common challenge in long-distance travel: balancing route efficiency with fuel cost optimization. By combining geolocation services, user-provided data, and intelligent distance calculations.

NLP-Powered-Amazon-Customer-Reviews-Analysis

This project applies Natural Language Processing (NLP) techniques to a large-scale Amazon customer review dataset (568,454 entries) to uncover actionable insights for business decision-making by combining structured data manipulation using SQL with unstructured text analysis using Python.

Fake-news-detection-ML-Model

In today's digital landscape, the spread of misinformation poses a serious threat to public trust, decision-making, and media credibility. This project leverages machine learning and natural language processing (NLP) to tackle the problem of fake news detection.

Income-Statement-Dashboard

This project focuses on using Power BI to visualize and analyze key financial metrics—**Revenue**, **Profit**, **Cost of Goods Sold (COGS)**, and **Discounts**—from financial data sourced from Power Bi covering multiple financial years and months.

Generative AI Medical Bot

I built a Generative AI–powered medical diagnosis platform designed to simulate real-world AI delivery constraints. The system uses FastAPI for backend services, LangChain for orchestration, Pinecone as a vector database, and Google Generative AI embeddings for semantic retrieval.

Computer (Food) Vision Project

With the growing popularity of food-related content in the digital space, having an efficient image classification model can greatly enhance food recognition applications. This project aims to build a deep learning-based system capable of classifying different types of food images accurately.

Automated_Webscraper

The Hacker News Web Scraper is a Python-based project that automates the extraction and analysis of top stories from Hacker News. Using BeautifulSoup and Requests, the scraper collects article titles, links, and upvotes, storing them in a structured Pandas DataFrame.

bank_marketing_campaign

This project analyzes a bank's marketing campaign data to understand customer behavior and predict term deposit subscription likelihood. The dataset contains information about customers' demographics, financial status, and campaign interactions.

Telco Customer Churn Prediction Model

Customer churn prediction is a crucial aspect for businesses to retain customers and improve profitability. This project aims to build a machine learning model to predict customer churn based on various features such as customer demographics, account information, and service usage patterns.

Predictive_modelling_of_pizza_sales

This project focuses on analyzing and predicting pizza sales using a regression model. The dataset captures various attributes of pizza sales over 2015, including pizza types, sizes, ingredients, and order times.

eagles wellness center

I recently participated in a group project that focused on creating a website for mental health. Our team collaborated to build an informative user-friendly platform that provides valuable resources and support for individuals dealing with mental health challenges.

Kiva_geospatial_project

In this project, I aim to investigate Kiva loans in the Philippines to identify regions that might be outside of Kiva's current network. Kiva reaches remote areas worldwide through its network of "Field Partners," which are local organizations that vet borrowers, provide services, and issue loans.

TimeSeries_forecasting_using_ARIMA_Model

Time series data is a sequence of data points collected or recorded at specific time intervals, such as monthly or daily. My goal is to understand the patterns in this data and make future predictions based on these patterns.

stock_prices_predictive_model

Stock Price Prediction Model Overview This project aims to predict the closing price of a stock using historical stock price data and machine learning techniques. The model utilizes features such as open, high, low, close prices, volume, 50-day moving average, and 200-day moving average to predict t

mt_cars_project

In this notebook, I will delve into the intricacies of multiple linear regression, building upon the foundational understanding of simple linear regression. We explore how to fit a model using sklearn, taking into account multiple predictor variables to predict a response variable.

supermarket_data_analysis

Through this project, I identified the most popular product lines, preferred customer payment methods, and the distribution of sales across different branches and customer types. These insights can be invaluable for businesses looking to optimize inventory and tailor marketing strategies

Experience

2019-2020

Data Analytics

Ardhi Sacco Society I was able to classify the data received into various departments in that all the records about those departments were easily linked to those departments. While viewing the data elements involved such as loan names in the given byproducts, I was able to use indexes to match the data in the byproducts with the data provided in the system. To do this I employed various tools like the v-lookup tool and the if statement analysis to get corresponding cell values. After the index elements, I derived reports of variances between the datasets by using a pivot table to analyze columnar data and then rearranging it into rows and filters to aggregate and group them into reports that can be easily presented and explained.

2022

Financial Reporting

Amp Bank Power Systems Ltd Here I set up an accounting system and reorganized their accounting framework from incomplete records while I was here I set up the manufacturing accounts, income statements, and their statement of financial position from which I was able to advise them on the steps they should take to improve their operations. I also conducted an audit and advised them on stricter methods they should take to curb such weaknesses.

2020-2023

Data Analyst

● Employed statistical methods and software to analyze data, identifying patterns and trends. ● Delivered clear and concise reports to stakeholders based on data interpretations, facilitating strategic decision-making. ● Communicated insights through reports, enhancing stakeholder decision-making. Resulted in a 15% increase in operational efficiency and a 10% improvement in strategic resource allocation.

Education

2012-2018

Financial Engineering at Jomo Kenyatta University of Agriculture & Technology

Financial Engineering refers to the application of mathematical, computational, and statistical techniques to design financial products, models, and algorithms to solve complex financial problems. The discipline has evolved over the past few decades, driven by the increasing complexity of financial markets and the growing demand for innovative financial products and solutions. The use of advanced analytical tools, such as stochastic calculus, optimization, and simulation, has become prevalent in the field of financial engineering. The discipline encompasses a variety of areas, including risk management, asset pricing, portfolio management, and financial product design. The goal of financial engineering is to create efficient and effective financial solutions that meet the investors needs.

2018-2019

Chartered Financial Analyst at Strathmore University

The CFA Level 1 curriculum provides a strong foundation in investment principles, financial analysis, and ethical practices. It covers topics such as quantitative methods, economics, financial reporting and analysis, equity investments, fixed income, derivatives, alternative investments, portfolio management, and corporate finance. The program emphasizes analytical skills, understanding of financial tools, and the application of ethical and professional standards in decision-making. It is designed to equip professionals with a comprehensive understanding of the global investment industry and the skills necessary to analyze and manage financial assets effectively.

2023-2024

Data Science at ALX

Data science is an interdisciplinary field that focuses on extracting knowledge and insights from raw data. It combines elements of statistics, computer science, and domain expertise to analyze large volumes of data and uncover patterns, trends, and insights. The goal of data science is to use data to drive decision-making and solve complex problems. Key techniques used in data science include data exploration and visualization, machine learning, statistical modeling, and data mining. In the age of big data, data science has become increasingly important across many industries, including finance, healthcare, marketing, and technology. Data scientists typically have a strong background in mathematics and statistics, as well as programming skills in languages such as Python or R.

2025 - 2026

Data Engineering at ALX

Data engineering is the discipline of building the architecture and infrastructure that allows organizations to manage and utilize data at scale. It focuses on creating reliable pipelines that move and transform data from diverse sources into formats that analysts, data scientists, and applications can use effectively.This work spans database design, ETL processes, workflow orchestration, and the use of distributed systems to handle large and complex datasets. Modern data engineering also emphasizes cloud-native solutions, real-time streaming, and scalable storage frameworks. Professionals in this field are skilled in programming, database technologies, and big data tools, and play a critical role in ensuring that data is trustworthy, accessible, and optimized for business needs.