Edge Computing and Edge Machine Learning

Introduction Data scientists often face three possibilities when deploying our products into production: through the cloud, edge computing, or the more recent variant, Edge Machine Learning. To introduce these concepts and provide context for this post, I will start by discussing the foundation—AI Computing. AI Computing and the Rigor of Mathematics A promise of computing is to make machines learn, and this is the job of the data scientist. When applying a model to data, Alan Turing’s insight from 1947 remains highly relevant. He simply stated that “machines learn from experience” (train set), a concept now popularized as data-driven methods. ...

October 14, 2024 · Carlos Daniel Jiménez

MLOps Guides: A Comprehensive Overview

Exploring the intersection of machine learning and DevOps - from model versioning to automated deployments. Featured Posts 📦 Artifact Design and Pipeline in MLOps Part I Introduction to artifacts, MLproject manifests, and pipeline orchestration for reproducible ML workflows. 🤖 MLflow for Generative AI Systems Learn how to use MLflow for tracing, evaluation, and versioning of LLM applications and Agentic AI systems. 🍓 Raspberry Pi 16GB, Servers, and MLOps Using Raspberry Pi as a development server for MLOps testing and edge deployments. ...

June 15, 2024 · Carlos Daniel Jiménez

📬 Did this help?

I write about MLOps, Edge AI, and making models work outside the lab. One email per month, max. No spam, no course pitches, just technical content.