Skip to content
View dendarko's full-sized avatar

Block or report dendarko

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 supported. This note will be visible to only you.
Report abuse

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

Report abuse
dendarko/README.md

Hi, I'm Dennis Darko 👋

I'm a Machine Learning & AI Engineer and MLOps Specialist based in North Vancouver, Canada, with over seven years of experience designing, deploying and maintaining scalable AI solutions. I specialize in generative AI, large language models and retrieval-augmented generation applications, as well as MLOps, data engineering and cloud-native deployments.

What I Do

Generative AI & NLP: Build and fine-tune large language models and transformer-based systems; design prompt engineering and retrieval-augmented generation (RAG) pipelines using LangChain and vector databases; create summarization and Q&A systems.

Machine Learning Engineering: Develop end-to-end ML pipelines from data ingestion and feature engineering to model training and evaluation; work with traditional ML and deep learning frameworks including scikit-learn, TensorFlow, PyTorch, CatBoost and DistilBERT.

MLOps & LLMOps: Implement CI/CD workflows, experiment tracking, model registry and data versioning using MLflow, DVC, Airflow, Vertex AI, Azure DevOps and GitHub Actions; deploy models across cloud platforms (Azure ML, AWS SageMaker, Google Vertex AI) and monitor them using Grafana, Prometheus and Cloud Logging.

Data Engineering: Design and maintain robust ETL/ELT pipelines on BigQuery, BigQuery, Dataflow and cloud functions; build and manage data warehouses, lakes and lakehouses; ensure data quality, scalability and reliability.

Project Leadership & Collaboration: Collaborate with cross-functional teams to align AI solutions with business goals; lead projects from conception to deployment; mentor team members and promote best practices in Agile development.

Technical Expertise

Category Tools & Technologies
Programming & Data Python, SQL, Bash, LookML, JavaScript
ML & Deep Learning scikit-learn, XGBoost, TensorFlow, PyTorch, CatBoost, OR-Tools, DistilBERT, Hugging Face, Keras, LangChain
Generative AI & NLP Large Language Models, transformer fine-tuning, prompt engineering, retrieval-augmented generation (RAG)
MLOps & Data Engineering MLflow, DVC, Airflow, Vertex AI, Azure ML, AWS SageMaker, Docker, Kubernetes, Bitbucket Pipelines, Azure DevOps, Git
Data Warehousing ETL/ELT, BigQuery, Redash, MySQL, PostgreSQL, Dataflow
Monitoring & Logging Grafana, Prometheus, Azure Monitor, Cloud Logging
Collaboration Tools Jira, Confluence, Slack, Looker, Tableau, Matplotlib, Plotly, Notion

Featured Projects

  • TransCostML – Command-line tool for transport price estimation; extracts delivery data, preprocesses it and trains ensemble models (Random Forest, XGBoost, stacking), achieving ~18% improvement in MAE. Repository
  • Gomat Markup Optimization – Conversion-probability and markup-optimization models using CatBoost and Random Forest; integrated FastAPI inference service with MLflow tracking and CI/CD pipelines; deployed on Azure ML Studio. Repository
  • GoSource Routing Optimization – Proof-of-concept route-optimization system using OR-Tools and Flask to generate optimal vehicle routes with CLI tools and Slack notifications. Repository
  • Ads Recommendation System – Real-time ads recommendation engine on GCP Vertex AI and BigQuery, providing personalized ads that improved click-through rates by 35%. Repository
  • Participedia Capstone – Multi-task learning pipeline using DistilBERT, deployed with Kubernetes and Vertex AI, generating embeddings and classifications for participatory democracy data. Repository
  • Loan Approval Prediction – Classification models (Logistic Regression, SVM, Random Forest, Gradient Boosting) to predict loan approvals, achieving an F1-score of 0.947. Repository

Resume

You can find my detailed resume here.

Contact

Pinned Loading

  1. gosource-routing-optimization gosource-routing-optimization Public

    Route optimization system that computes optimal vehicle routes among suppliers using OR-Tools; provides CLI to parse JSON inputs, QA scripts and Slack notifications; containerized and integrated wi…

    1

  2. ads-recommendation-system ads-recommendation-system Public

    Real-time ads recommendation engine built using GCP Vertex AI and BigQuery, delivering personalised ad recommendations with 35% higher click-through rates.

  3. gomat-markup-optimization gomat-markup-optimization Public

    Markup optimization system that trains conversion and markup optimization models using CatBoost and Random Forest; provides a FastAPI inference API with MLflow tracking; containerized and deployed …

  4. loan-approval-prediction loan-approval-prediction Public

    Loan approval prediction project using scikit-learn classification models (logistic regression, SVM, decision tree, random forest, gradient boosting), achieving an F1 score of 0.947.

  5. participedia-capstone participedia-capstone Public

    Participatory democracy analysis project using DistilBERT and multi-task learning, deployed with Vertex AI, Kubernetes and DVC.

  6. transcostml transcostml Public

    CLI pipeline for transportation cost estimation with ensemble and linear regression models (Random Forest, XGBoost, stacking); includes ETL, preprocessing, training, hyperparameter tuning and MLflo…