An end-to-end machine learning project predicting DoorDash delivery durations, utilizing MLOps principles and best practices. This project is currently in progress.

This project implements a Retrieval-Augmented Generation (RAG) pipeline using Mage to manage large language models (LLMs) on an hourly schedule.

This Jupyter notebook project evaluates the accuracy of language model responses to generated questions by comparing them to a ground truth dataset using cosine similarity and ROUGE metrics.

This project demonstrates using `Elasticsearch` and vector search techniques to efficiently find answers to user questions in FAQ documents by leveraging embeddings and evaluating search performanc…

This project integrates the OpenAI API with a locally running Docker container to generate text completions using the gemma:2b model, as demonstrated in the Jupyter Notebook `ollama.ipynb`.

This repository retrieves and indexes course-related FAQ documents in Elasticsearch for advanced querying and constructs detailed prompts to generate contextually relevant answers using OpenAI's mo…

DataTalks: MLOps Zoomcamp - Homework for Module 5 (Monitoring)

DataTalks: MLOps Zoomcamp - Homework for Module 4

DataTalks: MLOps Zoomcamp - Homework for Module 3

DataTalks: MLOps Zoomcamp - Homework for Module 2

DataTalks: MLOps Zoomcamp - Homework for Module 1