Looking for a job

Nyasha Makuto

@nyashamakuto

AI-focused data scientist building predictive models and agentic LLM workflows.

Data scientist with 5 years of experience in predictive modeling, machine learning, statistical modeling, hypothesis testing, and general analytics in Python and R. Skilled in working on imbalanced classification problems with deep learning models in Tensorflow and PyTorch, as well as with traditional machine learning models such as XGboost, random forests, logistic regression, and support vector machines (SVMs). Also skilled in unsupervised methods such as principal component analysis (PCA) and clustering.

Experience

Work history, roles, and key accomplishments

Current

AI Scientist, Advanced Analytics and AI

Current

Canadian Imperial Bank of Commerce

Oct 2022 - Present (3 years 6 months)

Built predictive models using Python and KQL, including an XGBoost classifier for malicious command line activity (43.2% test set PR AUC) and a GPU-enabled CNN for malicious URL detection (99.85% test set PR AUC). Developed an agentic LLM workflow to summarize, classify, and respond to customer emails and used Langfuse to monitor prompt/agent performance.

Python PyTorch XGBoost Langfuse Generative AI CNN Anomaly Detection

Model Risk Specialist (Model Validation)

Bank of Montreal (BMO)

Jun 2021 - Sep 2022 (1 year 3 months)

Validated and challenged financial fraud and anti-money laundering models by reviewing model design choices across classifiers, tree-based models, ensemble methods, and neural network approaches. Replicated and reviewed Python, SAS, and Spark SQL implementations as well as spreadsheet calculations.

Model Validation Anti Money Laundering (AML)Fraud Models Python SAS BERT Vectorizers Neural Networks

Junior Data Scientist

Preteckt

Feb 2021 - Apr 2021 (2 months)

Queried large datasets from cloud environments using SQL in Jupyter notebooks and JupyterHub, including TimescaleDB and DigitalOcean. Developed in-house Python modules and version-controlled contributions in Git with multiple co-authors.

SQL TimeScaleDB DigitalOcean Jupyter Notebooks JupyterHub Python Git

Graduate Assistant (Data Analyst)

Lakehead University

Sep 2018 - Apr 2020 (1 year 7 months)

Used Stata to conduct statistical analyses on mental health data and presented findings at academic conferences in Canada and the U.S. Applied rigorous statistical methods to interpret results and communicate insights to technical audiences.

Stata Statistical Analysis Hypothesis Testing Regression Modeling