7+

Years Experience

20+

ML Projects

6

Featured Projects

100%

Production Focus

About Me


Hi! I'm Israel Igietsemhe, a Lead Data Scientist with over seven years of experience building production-ready machine learning systems and AI solutions. I specialize in end-to-end ML engineering, from model development to deployment, with expertise in MLOps, agentic AI, and cloud infrastructure.

My work spans multiple domains including supply chain optimization, healthcare analytics, recommender systems, and natural language processing. I'm passionate about creating scalable, maintainable ML systems that solve real-world business challenges. Currently, I lead data science initiatives at KetteQ, a SaaS company focused on supply chain optimization, where I develop and deploy forecasting models that drive operational efficiency.

Beyond traditional ML, I'm deeply involved in cutting-edge AI technologies including agentic AI systems, RAG (Retrieval-Augmented Generation) architectures, and multi-agent orchestration. I combine strong technical skills in Python, cloud platforms (Azure, GCP), and modern MLOps tools (MLflow, DVC, Airflow) with a focus on production deployment and infrastructure as code.

I believe in the power of data-driven decision making and enjoy working in collaborative teams to tackle challenging problems. Whether it's building a recommender system with full MLOps practices, deploying agentic AI for healthcare applications, or architecting cloud-native ML pipelines, I bring a holistic approach to every project.

I love working in exciting teams solving challenging problems!

Download Resume

Work Experience


KetteQ Sep 2022 - Present

Lead Data Scientist

  • Responsible for the development and deployment of machine learning forecasting models on the KetteQ demand and operations planning software platform.
  • Afrilearn May 2022 - Present

    Machine Learning Consultant

  • Participating in a $100,000 UNICEF-sponsored project for making education more accessible to indigent students across Africa. Deploying a recommender system for an e-learning platform using an association rule mining model. Applying Open AI ChatGPT for generating multiple choice questions for quiz games.
  • Loblaw Companies Limited Sep 2021 - April 2022

    Data Science Analyst, Healthcare Data Products and Services

  • Implemented a boosted tree classifier on Big Query ML for predicting patient churn with an average accuracy of 65%.
  • Initiated and implemented a patient likelihood classification model for predicting the most likely patients to respond to a recommendation notification with about 70% testing accuracy.
  • Researched and presented concepts on the mathematics of deep learning and recommender systems to my team members on a biweekly basis.
  • Sharpest Minds April 2021 - Aug 2021

    Data Science Fellow

  • Built a full-stack machine learning web application that classifies Amazon product reviews using natural language processing. The product helps customers clarify product selection based on keywords discussed in product reviews.
  • Created an original dataset by building a data pipeline that scrapes Amazon product reviews using beautiful soup and performs data cleaning to prepare the dataset for topic modeling.
  • Applied natural language processing topic modeling using a latent dirichlet allocation model in scikit-learn to identify keywords for each product review.
  • Deployed model as a web-app using Python Flask and Heroku.
  • University of Toronto Sep 2016 - Aug 2021

    Machine Learning Team Lead, Dynamic Optimization and Operations Management LAB

  • Collected user and business requirements and put together detailed project plans.
  • Developed random forest models in helping Nestle Canada in underpayment claims classification.
  • Developed a local heuristic search model for inventory allocation for Nestle Canada with potential reduction in penalty costs up to $200,000 monthly.
  • Developed demand forecasting models using multivariate regression models, generative adversarial networks and Bayesian LSTMs using keras, sk-learn, Tensorflow and PyTorch.
  • CIBC Jul 2017-Aug 2017

    Machine Learning Specialist

  • Initiated an unsupervised learning system for anomaly detections in managing role-based access control using hierarchical clustering of the jaccard distances.
  • Implemented python codes using selenium package for web scraping of employee profiles.
  • Featured Projects


    A collection of production-ready machine learning and data science projects showcasing end-to-end MLOps practices, advanced algorithms, and real-world applications.

    E-commerce Recommender MLOps

    A production-style end-to-end recommender system for e-commerce platforms with data versioning, experiment tracking, drift monitoring, and deployment capabilities.

    Technologies:

    Python · DVC · MLflow · Evidently · Implicit · FastAPI/Streamlit · GitHub Actions

    Key Features:

    • Data & model versioning with DVC
    • Experiment tracking & model registry (MLflow)
    • Data & concept drift detection (Evidently)
    • Deployable API/UI with FastAPI/Streamlit
    • CI/CD automation with GitHub Actions
    View on GitHub

    Agentic Clinical Assistant

    A production-ready, regulated-domain-safe agentic RAG system that orchestrates multiple specialized agents to provide clinical operations guidance with strict safety, grounding, and auditability requirements.

    Technologies:

    Python · RAG · LLM · Multi-Agent Systems · Healthcare AI · Safety & Compliance

    Key Features:

    • Multi-agent orchestration for clinical operations
    • Regulated-domain safety & compliance
    • Retrieval-Augmented Generation (RAG)
    • Strict grounding and auditability
    • Production-ready deployment
    View on GitHub

    Supply Chain Airflow ML

    An end-to-end machine learning pipeline for supply chain optimization using Apache Airflow for workflow orchestration, featuring demand forecasting, inventory optimization, and automated ML workflows.

    Technologies:

    Python · Apache Airflow · Machine Learning · Time Series Forecasting · Supply Chain Analytics

    Key Features:

    • Automated ML pipeline orchestration
    • Demand forecasting models
    • Inventory optimization algorithms
    • Workflow scheduling & monitoring
    • Scalable data processing
    View on GitHub

    Graph Neural Networks

    Research and implementation of Graph Neural Network (GNN) architectures for learning representations on graph-structured data, including node classification, link prediction, and graph-level tasks.

    Technologies:

    Python · PyTorch Geometric · Deep Learning · Graph Theory · Neural Networks

    Key Features:

    • GNN architecture implementations
    • Node & graph-level learning
    • Link prediction models
    • Graph representation learning
    • Research & experimentation framework
    View on GitHub

    Ad Campaign Optimizer

    A machine learning system for optimizing digital advertising campaigns through budget allocation, audience targeting, and performance prediction to maximize ROI and campaign effectiveness.

    Technologies:

    Python · Machine Learning · Optimization Algorithms · Marketing Analytics · Data Science

    Key Features:

    • Budget allocation optimization
    • Audience targeting models
    • Campaign performance prediction
    • ROI maximization algorithms
    • Real-time campaign analytics
    View on GitHub

    Azure GenAI

    A FastAPI-based application that integrates Azure OpenAI (GPT-4o) and Azure AI Search to provide chat, embedding, and RAG (Retrieval Augmented Generation) capabilities with vector search for semantic document retrieval.

    Technologies:

    Python · FastAPI · Azure OpenAI · Azure AI Search · GPT-4o · RAG · Vector Search

    Key Features:

    • Chat API with GPT-4o integration
    • RAG-powered Q&A with document search
    • Embedding generation API
    • Azure AI Search vector search capabilities
    • Production-ready FastAPI deployment
    View on GitHub

    Skills


    I am equipped with diverse skills by the virtue of my hard work. In the past, I have worked on multiple projects in churn prediction, natural language processing, demand forecasting, supply chain optimization, recommender systems, MLOps, graph neural networks, and LLM/RAG systems. Explore my website to get more details about my experience and projects.

    Programming Languages

    Python

    99%

    R

    99%

    Machine Learning & AI

    Machine Learning

    99%

    Natural Language Processing/GPT

    95%

    Recommender Systems

    92%

    Graph Neural Networks

    85%

    RAG & LLM Systems

    88%

    Time Series Forecasting

    95%

    Optimization Algorithms

    90%

    MLOps

    MLOps (MLflow/DVC/Evidently)

    90%

    Apache Airflow

    85%

    Terraform

    80%

    Cloud Platforms

    Google Cloud Platform/Big Query ML

    90%

    Microsoft Azure

    80%

    Data & Analytics

    SQL

    95%

    Pyspark

    80%

    DevOps & Deployment

    Docker/Flask/Model deployment

    90%

    FastAPI/Streamlit

    90%

    CI/CD (GitHub Actions)

    85%

    Git/GitHub/GitLab/Bitbucket

    95%

    Kubernetes

    70%

    Contact


    Thank you for going through my portfolio, don't hesitate to send me an email at aloagbaye.i@gmail.com