Data Scientist · themu.co

Mustafa
Ali Mohammed

I find patterns where others see noise — turning raw data into decisions through machine learning, statistical modelling, and clear thinking.

µ
Population Mean
Gradient Descent
Statistical Models
→0
Loss Minimised
Data Science· Machine Learning· Statistical Modelling· themu.co· Python· Neural Networks· Data Visualisation· Mustafa Ali Mohammed· Data Science· Machine Learning· Statistical Modelling· themu.co· Python· Neural Networks· Data Visualisation· Mustafa Ali Mohammed·
About
01

I make machines
understand the world.

I'm Mustafa — a data scientist working at the boundary between raw data and human meaning. I move across the full stack: exploratory analysis, feature engineering, model deployment, and communicating results to people who don't speak Python.

The µ in my brand isn't just my domain — it's the population mean, the parameter every data scientist is trying to estimate. You never have perfect information, but you can get remarkably close.

Python
Core Language
Machine Learning
Sklearn · PyTorch
Statistics
Inference · Modelling
SQL
Data Engineering
Visualisation
Matplotlib · D3
Deep Learning
NLP · Vision
Selected Work

Projects
& Research

01
Predictive Churn Modelling
End-to-end churn prediction pipeline for a subscription business. Gradient boosted trees with SHAP explainability reduced false-positive rate by 34% over baseline logistic regression.
PythonXGBoostSHAPClassification
02
NLP Sentiment Pipeline
Fine-tuned BERT on domain-specific customer reviews to extract granular sentiment across product dimensions. Deployed as a REST API processing 50K records/day.
NLPBERTHuggingFaceFastAPI
03
Time Series Anomaly Detection
Unsupervised anomaly detection for IoT sensor data using Isolation Forest and LSTM autoencoders. Flagged equipment failures 72 hours before downtime in production.
Time SeriesLSTMAnomaly DetectionPyTorch
04
Bayesian A/B Testing Framework
Reusable Bayesian A/B testing framework for product teams, replacing frequentist t-tests with posterior probability estimation and early stopping rules.
Bayesian StatsExperimentationPython
Experience
03
2024 — Now
Data Scientist
Independent · Freelance
Working with startups and scale-ups to build data science capabilities — from first model to production pipeline.
2022 — 2024
Junior Data Analyst
Previous Role
Built dashboards, automated reporting pipelines, and ran statistical analyses to support product decisions.
2019 — 2022
BSc Data Science
University
Foundations in statistics, machine learning, algorithms, and applied mathematics.

Let's
build
something.

Open to freelance projects, full-time roles, and interesting conversations about data.