EPAEmmanuel Paddy Adams
Résumé

Open to ML & data roles · on-site in Accra or remote

Emmanuel Paddy Adams

I am a Data Scientist and Machine Learning Engineer.

I build machine learning models end to end: collecting and cleaning real-world data, engineering features, and evaluating models honestly. Then I get them out of the notebook and into tools people can use.

Read the featured case study Download CV
  • Accra, Ghana
  • Currently building: Customer Churn Prediction
Portrait of Emmanuel Paddy Adams
R² · Ghana Maize Yield Prediction
0.95Random 80/20 split; likely optimistic
R² · Used Car Price Prediction
0.875-fold cross-validation
Projects published
7
Data certifications
3

About

Statistics first, engineering close behind.

I’m a Data Scientist and machine learning engineer in Accra, Ghana, with a degree in Statistics with Computer Science from the University of Ghana. That grounding shapes how I work. I care as much about how a model is evaluated as how it’s built, and I’d rather report an honest number than an impressive one.

I’ve built regression and classification models on agricultural, automotive and telecom data. They include a district-level maize yield model that combines Ghanaian yield records with NASA satellite weather data. I’ve also built analytics dashboards, including one on a 286-person survey of Ghana’s tech workforce, and worked with insurance data as a data analyst intern at Unique Insurance.

My current focus is the engineering side of ML: reproducible pipelines, time-aware validation, and serving models behind simple apps and APIs.

  • Regression & classification models
  • Feature engineering & validation
  • scikit-learn pipelines & model packaging
  • Analytics & BI dashboards

Outside of work: football, reading, music, gaming and content creation.

Selected work

Featured projects

All projects →
Data Science & Machine Learning0.95 R²

Ghana Maize Yield Prediction

An end-to-end pipeline predicting district-level maize yields in Ghana. It combines yield records with growing-season weather from the NASA POWER API, soil types and policy/pest indicators, then trains a Random Forest + Gradient Boosting ensemble.

  • Python
  • scikit-learn
  • Pandas
  • NASA POWER API
  • Ensembles
Approach & impact

Approach

Standardised district names to merge yield, weather and soil sources; aggregated daily NASA weather into April–August growing-season features; added a previous-year yield lag, pest-outbreak and Planting for Food and Jobs policy indicators; trained a voting ensemble inside a scikit-learn pipeline.

Why it matters

District-level yield forecasts can support planning by farmers, agencies and food-security programmes in Ghana.

Read case study →Code on GitHub
Data Science & Machine Learning0.87 R²

Used Car Price Prediction

Regression models predicting car prices for ~11,900 vehicles from engine specs, fuel type, transmission, body style and year. Compares OLS, Lasso and Ridge, with Lasso used for feature selection.

  • Python
  • scikit-learn
  • Lasso / Ridge
  • Feature engineering
Approach & impact

Approach

Started from a numeric-only baseline (test R² 0.57), then added one-hot encoded categoricals with rare-model grouping, raising test R² to 0.82. A cross-validated Lasso kept 86 of 182 features; Ridge reached R² 0.87 under 5-fold cross-validation.

Why it matters

A full regression workflow on messy tabular data, covering baselining, feature engineering, regularisation and honest validation.

Code on GitHub
Analysis286 Respondents

Ghana Tech Ecosystem Survey 2025

Turned raw survey responses from 286 Ghanaian tech professionals into a 4-page interactive Power BI dashboard covering roles, skills gaps, AI tool adoption, income and ecosystem needs.

  • Python
  • Power BI
  • DAX
  • Survey analysis
Approach & impact

Approach

Cleaned and de-duplicated responses (279 unique respondents), reshaped 43 questions into a 7,425-row long format, and built drill-through dashboard pages with 18 calculated measures.

Why it matters

Findings such as fintech accounting for 41% of respondents, an 11.5% female share and better-paying local jobs as the top need (56%) give policymakers, investors and trainers a clear evidence base.

Code on GitHub

Experience

Experience & education

Work

  1. Mar 2026

    Data Analyst Simulation

    BCG · Forage virtual programme · Remote

    • Analysed multi-channel advertising campaign data to evaluate conversion rates and ROI across audience segments.
    • Identified the highest-performing channels and delivered a recommendation deck with data-backed marketing strategy.
    View project
  2. Nov 2025 – Dec 2025

    Data Analyst Intern

    Unique Insurance · Accra, Ghana

    • Validated and audited insurance datasets for 3 internal teams, flagging inconsistencies across 1,000+ records to reduce reporting discrepancies.
    • Built ad-hoc reports and data organisation templates adopted by the audit team for recurring monthly use.
    • Worked within data quality controls and compliance-driven reporting in a regulated financial services environment.

Education

  1. 2026

    BSc Statistics with Computer Science

    University of Ghana

    Advanced Linear Regression, Machine Learning, Database Systems & Management, Theory of Sampling

Training & certifications

  • DataBloom Africa · Data Analysis Programme · Cohort 12025
  • ALX Africa · Professional Foundation & Data Analytics2024 – 2025
  • Cognitive Class (IBM) · Data Analysis with Python2025

Skills

What I do

Machine learning

Regression and classification with linear models (OLS, Lasso, Ridge) and tree ensembles (Random Forest, Gradient Boosting, XGBoost). I handle class imbalance, tune with cross-validation, and pick metrics that match the business cost of errors.

  • scikit-learn
  • XGBoost
  • SMOTE
  • Cross-validation

Data pipelines

Collecting data from APIs and messy files, joining sources reliably, and engineering features (lags, encodings, aggregations) inside reproducible scikit-learn pipelines that can be saved and served.

  • Python
  • Pandas
  • NumPy
  • SQL

Analytics & communication

Exploratory analysis and interactive dashboards that turn results into decisions for non-technical stakeholders, plus clear write-ups of what a model can and can't do.

  • Power BI
  • DAX
  • Matplotlib
  • Seaborn

Toolkit

Languages

Python, SQL

ML & data

scikit-learn, XGBoost, Pandas, NumPy

Databases & BI

MySQL, PostgreSQL, Power BI, Excel

Workflow

Git, GitHub, Jupyter, VS Code

Contact

Let's work together.

I’m looking for machine learning and data roles, and I’m open to freelance and research collaborations. The quickest way to reach me is by email.