Skip to content
View dmortalla's full-sized avatar

Block or report dmortalla

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
dmortalla/README.md

👋 Hi, I'm Darrell Mortalla

AI Engineer | LLMs · RAG · Machine Learning · Data Systems

I build production-style AI systems, including LLM applications, RAG pipelines, and end-to-end ML workflows.

My work focuses on combining LLMs, machine learning, and data engineering into real-world, production-ready applications.

My work spans:

  • LLM applications (RAG pipelines, tool-augmented agents, conversational memory)
  • Machine learning (PyTorch, Scikit-learn)
  • ML engineering (FastAPI, MLflow, Docker)
  • Data systems (PySpark, DuckDB, cloud pipelines)

🧩 Featured Projects

🚀 AI & Machine Learning Projects (Flagship)

🤖 Personal AI Assistant — RAG + Memory + Tools

Production-style AI assistant demonstrating LLM orchestration, retrieval pipelines, and tool-augmented reasoning.

Built with Streamlit featuring:

  • Retrieval-Augmented Generation (RAG)
  • Tool execution (calculator via AST parsing)
  • Conversational memory
  • Modular architecture

Tech: Python, Streamlit, LLMs

💼 Use Case: AI assistant capable of answering questions from documents, performing calculations, and maintaining conversational context.

🧱 Architecture: Modular pipeline with retrieval layer, tool execution layer, and memory management


🧠 ML-Powered Recommendation System (Dog Breed Assistant)

Machine learning–based recommendation system for dog breed selection, deployed as an interactive Streamlit application.


⚙️ ML Pipeline — Churn Prediction (MLflow + FastAPI)

Production-style ML system covering the full lifecycle:

  • Model training
  • Experiment tracking (MLflow)
  • API deployment (FastAPI)

🏗️ Data Engineering Pipeline (Lakehouse)

Medallion architecture pipeline using DuckDB and modular ETL design.

🔍 Focus Areas: LLM applications, RAG systems, ML pipelines, and AI system design & architecture


📊 Data Analytics & BI Projects (Foundation)

A collection of analytics projects covering data cleaning, visualization, and business intelligence.

# Project Description Tools
1 Data Cleaning & Preparation Retail Sales dashboard with AWS deployment — end-to-end data wrangling workflow with validation scripts. Python · Pandas · AWS
2 Exploratory Data Analysis (EDA) Customer Churn tracker deployed on Azure — statistical profiling and churn behavior analysis. Python · Matplotlib · Azure
3 Excel Automation Dashboard COVID-19 trend analysis with GCP integration — automated Excel reports powered by Python scripting. Python · Excel · GCP
4 Power BI Retail Insights Marketing Campaign ROI analytics — cross-channel marketing performance with Power BI dashboards. Power BI · SQL
5 Tableau Global Sales Overview E-commerce order automation using Python + Excel — visual storytelling with Tableau and automated ETL. Tableau · Python
6 SQL Sales Optimization Financial KPI tracking via SQL + Power BI — optimized queries for pricing and profitability insights. SQL · Power BI
7 Predictive Modeling Customer segmentation & RFM analysis with Tableau — machine-learning-based forecasting and insights. Python · Tableau
8 Customer Segmentation Supply Chain efficiency modeling with Power BI — advanced segmentation and dashboard optimization. Power BI · SQL
9 A/B Testing & Statistical Analysis Marketing campaign inference modeling — statistical testing to identify significant strategies. Python · Statsmodels
10 Sales Forecasting Time-series prediction using Python + Power BI — ARIMA and Prophet models for trend forecasting. Python · Power BI
11 BI Data Modeling Dimensional warehouse schema (SQL + Power BI) — data modeling and ETL processes for BI systems. SQL · Power BI
12 Executive Insights Dashboard Storytelling dashboard for business leaders — executive-level insights visualized in Power BI. Power BI
13 Cloud Data Pipeline Capstone End-to-end analytics pipeline on AWS + Power BI — full-stack ETL, storage, and BI pipeline. Python · AWS · Power BI

🛠️ Tools & Technologies

Python SQL Excel Power BI Tableau AWS Azure GCP


📫 Connect With Me


“Data transforms decisions into measurable success.”
© 2025 Darrell Mortalla

Add portfolio overview README

Pinned Loading

  1. urban-mobility-data-lakehouse urban-mobility-data-lakehouse Public

    Production-style medallion lakehouse pipeline for NYC taxi data built with Python, Pandas, Parquet storage, and DuckDB for analytics, demonstrating modular ETL design and reproducible data engineer…

    Python

  2. dog-breed-chatbot dog-breed-chatbot Public

    AI-powered dog breed recommender with Streamlit.

    Python

  3. mlai-portfolio mlai-portfolio Public

    Machine Learning & AI Engineering Portfolio — Interactive Apps, RAG, Deep Learning, Recommenders, Forecasting, and Anomaly Detection.

    Python

  4. transformer-from-scratch-pytorch transformer-from-scratch-pytorch Public

    A clean, educational PyTorch Transformer encoder implementation built from first principles.

    Python