Skip to content
View kachiann's full-sized avatar
💭
🔬 Researching QMC methods
💭
🔬 Researching QMC methods

Block or report kachiann

Block user

Prevent this user from interacting with your repositories and sending you notifications. Learn more about blocking users.

You must be logged in to block users.

Maximum 250 characters. Please don’t include any personal information such as legal names or email addresses. Markdown is supported. This note will only be visible to you.
Report abuse

Contact GitHub support about this user’s behavior. Learn more about reporting abuse.

Report abuse
kachiann/README.md

Hi, I'm Kachi 👋

Data Engineer & ML/AI Practitioner, PhD — production ELT pipelines, applied ML systems, and a research background in computational mathematics.

I design and build data systems that hold up in production — layered ELT pipelines, cloud data warehouses, and applied ML/AI systems — with the same rigor I spent five years applying to peer-reviewed numerical research.

  • 🔭 Building cloud-native ELT pipelines and MLOps systems (GCP · dbt-style layered warehouses · Prefect · MLflow)
  • 🤖 Shipping RAG and LLM-powered applications grounded in real data sources, not model guesswork
  • 🎓 Lecturing in Data Analytics & Machine Learning (Berlin School of Business & Innovation)
  • 📄 PhD in Computational Mathematics (Quasi-Monte Carlo methods) · 5 peer-reviewed publications · Google Scholar · ORCID
  • 📝 Writing on Medium

🛠 Tech Stack

Languages

Python SQL R

Data Engineering & Cloud

GCP Terraform Docker Spark

ML & MLOps

PyTorch Scikit-Learn Pandas MLflow

LLM / NLP

LangChain HuggingFace


📌 Featured Projects

Project Description Stack
Maternal Health AI Assistant Agentic RAG assistant answering infant care questions — grounded in WHO/AAP/CDC guidelines and live USDA nutrition data, not model memory LangChain · GPT-4o · FAISS · Streamlit
NourishMama Cloud-native ELT pipeline and layered BigQuery warehouse (raw → staging → marts) powering a nutrition dashboard GCP · Terraform · Bruin · Streamlit
Immobilien Preisrechner RLP Regional house price model (R² 0.75) with SHAP explainability and MLflow-tracked experiments XGBoost · SHAP · MLflow · Streamlit
MLOps Bike Demand End-to-end MLOps pipeline with orchestration, CI/CD, and automated drift monitoring MLflow · Prefect · Docker · Evidently
Hotel Booking Cancellations ML model predicting cancellations to support dynamic pricing, retention targeting, and overbooking optimization Python · Scikit-Learn · Gradio

More case studies (with problem/approach/outcome write-ups): kachiann.github.io/portfolio-v2


🤝 Connect

LinkedIn Medium X Google Scholar

🟢 Open to full-time and contract roles in data engineering, ML engineering, and analytics — based in Frankfurt, Germany (EU work authorisation). DM on LinkedIn or reach me at kachi.emenike12@gmail.com.

Pinned Loading

  1. maternal-health-ai-assistant maternal-health-ai-assistant Public

    Agentic AI assistant for nursing mothers. RAG over WHO/AAP/CDC sources combined with live USDA nutrition API tool-use. Built with LangChain, GPT-4o, FAISS, and Streamlit.

    Python 1

  2. immo-preisrechner-rlp_de immo-preisrechner-rlp_de Public

    End-to-end house price predictor for Rhineland-Palatinate using XGBoost, MLflow, Streamlit, and SHAP

    Python

  3. Citi-Bike-Analytics-Pipeline Citi-Bike-Analytics-Pipeline Public

    Batch data engineering pipeline on GCP (Terraform, GCS, BigQuery) with partitioned warehouse and interactive Streamlit dashboard.

    Python 1

  4. project-mlops project-mlops Public

    Predict bike-sharing demand using machine learning pipeline for MLOps-Zoomcamp project, optimizing bike distribution and availability.

    Jupyter Notebook 2

  5. Bike_Sharing_Streamlit_App Bike_Sharing_Streamlit_App Public

    A Streamlit web application for analyzing bike sharing data and predicting rental demand using machine learning. Features include data exploration, usage pattern visualization, and demand forecasti…

    Jupyter Notebook

  6. financial-time-series-explorer financial-time-series-explorer Public

    Financial Time Series Explorer – a Streamlit app to visualize prices, returns, volatility, drawdowns, and correlations for market instruments.

    Python