Qualcomm® AI Hub helps to optimize, validate, and deploy machine learning models on-device for vision, audio, and speech use cases. With Qualcomm® AI Hub, you can: Convert trained models from frameworks like PyTorch and ONNX for optimized on-device performance on Qualcomm® devices. Profile models on-device to obtain detailed metrics including runtime, load time, and compute unit utilization. Verify numerical correctness by performing on-device inference. Easily deploy models using Qualcomm® AI Engine Direct, TensorFlow Lite, or ONNX Runtime. The Qualcomm® AI Hub Models repository contains a collection of example models that use Qualcomm® AI Hub to optimize, validate, and deploy models on Qualcomm® devices. Qualcomm® AI Hub automatically handles model translation from source framework to device runtime, applying hardware-aware optimizations, and performs physical performance/numerical validation. The system automatically provisions devices in the cloud for on-device profiling and inference. The following image shows the steps taken to analyze a model using Qualcomm® AI Hub.

This blog compares three popular machine learning workflow orchestration tools: ZenML, Flyte, and Metaflow. It explores their features, use cases, and strengths, helping data scientists and engineers choose the best option for building and managing efficient ML pipelines.

Demetrios chats with Zach Wallace, engineering manager at Nearpod, about integrating AI agents in e-commerce and edtech. They discuss using agents for personalized user targeting, adapting AI models with real-time data, and ensuring efficiency through clear task definitions. Zach shares how Nearpod streamlined data integration with tools like Redshift and DBT, enabling real-time updates. The conversation covers challenges like maintaining AI in production, handling high-quality data, and meeting regulatory standards. Zach also highlights the cost-efficiency framework for deploying and decommissioning agents and the transformative potential of LLMs in education.

//Abstract AI has transformed industries, yet its true potential often lies untapped within core business processes. In this session, we’ll explore how AI agents differ from generative AI models, emphasizing their deterministic, hallucination-free approach to problem-solving. We’ll take a live example of an AI Agent in the logistics sector, and will detail the architectural foundations that enable AI agents to reason effectively, execute chain-of-thought workflows, and integrate seamlessly into human teams. We’ll discuss how these agents confidently navigate complex, multimodal tasks, extracting structured insights from unstructured data, and leveraging dynamic workflows for maximum flexibility. With customizable confidence thresholds, statefulness to track long-term cases, and advanced document understanding, these agents solve real business challenges, such as processing autonomously claims till resolution, with precision. Through a live case study, we’ll illustrate the measurable top and bottom-line effects of deploying AI agents—highlighting significant efficiency gains, multilingual capabilities, and safe, scalable applications in mission-critical environments. By showcasing how AI agents mimic human decision-making at unparalleled speed, we’ll inspire senior management to rethink AI’s role in their organizations and harness its full potential for transformative impact. //Bio Passionate about connecting deep tech to end-users, Vanessa’s work is at the forefront of AI’s transformative potential. For over a decade, she has been transforming cutting-edge innovations into actionable solutions that drive industry change. This is a bi-weekly "Agent Hour" event to continue the conversation about AI agents. Thanks to arcade-ai.com for the support! Join the next live event at home.mlops.community

Demetrios chats with Egor Kraev, principal AI scientist at Wise, about integrating LLMs to enhance ML pipelines and humanize data interactions. Egor discusses his open-source MotleyCrew framework, career journey, and insights into AI's role in fintech, highlighting its potential to streamline operations and transform organizations.

This blog explores Retrieval-Augmented Generation (RAG), a technique that enhances large language models (LLMs) by incorporating external data to provide more accurate, up-to-date, and context-specific responses. The author, Steven Wu, shares insights from a year of building RAG-based applications and explains the concept in a simple analogy for beginners.

In this blog post, Dr Gleb Lukicov, discusses the setup for fast local development and scalable remote deployment of ML projects using pipelines, particularly when working with LLMs. Gleb shares how to implement local testing using Kubeflow Pipelines to shorten the development cycle using Docker cache, multi-stage builds, uv and dynamic user credentials injection. The full end-to-end implementation is available as a self-contained repository, a companion to this blog post, that also includes some infrastructure goodies like GitHub CI/CD & pre-commit config for linting & testing, local scripts with typer, project dependency management with uv, and static checking with mypy.

Part 1 of YouGot.us’ AI Agent study, conducted during the AI Agent in Production Event by Prosus and MLOps.Community, covers the definition of AI Agents, users, and use cases, as well as agent adoption.