81 posts with this tag
Meet Kitaru — open source durable execution for Python agents, built by the ZenML team. Crash recovery, human-in-the-loop, and replay from any checkpoint.
I rebuilt zenml.io — 2,224 pages, 20 CMS collections — from Webflow to Astro in a week using Claude Code and a multi-model AI workflow. Here's how.
ZenML's new pipeline deployments feature lets you use the same pipeline syntax to run both batch ML training jobs and deploy real-time AI agents or inference APIs, with seamless local-to-cloud deployment via a unified deployer stack component.
ZenML launches Pipeline Deployments, a new feature that transforms any ML pipeline or AI agent into a persistent, high-performance HTTP service with no cold starts and full observability.
FloraCast is a production-ready template that shows how to build a forecasting platform—config-driven experiments, model versioning/staging, batch inference, and scheduled retrains—with ZenML and Darts.
Discover why production teams are treating agentic workflows as MLOps evolution, not revolution—plus how ZenML achieved 200x performance improvements for enterprise ML operations. Real insights from 130+ MLOps engineers on building reliable AI systems.
A technical deep dive into the performance optimizations that improved ZenML's throughput by 200x
Our monthly roundup: new features with 0.80.0 release, more new models, and an MCP server for ZenML
ZenML 0.80.0 transforms tenant structures into workspace/project hierarchies with advanced RBAC for Pro users, while enhancing tagging, resource filtering, and dashboard design. Open-source improvements include Kubernetes security upgrades, SkyPilot integration, and significantly faster CLI operations. Both Pro and OSS users benefit from dramatic performance optimizations, GitLab improvements, and enhanced build tracking.
The OpenPipe integration in ZenML bridges the complexity of large language model fine-tuning, enabling enterprises to create tailored AI solutions with unprecedented ease and reproducibility.
Our monthly roundup: Hamza visits the US, a new course built on ZenML and why workflows are better than autonomous agents!
ZenML 0.75.0 introduces dashboard enhancements that allow users to create and update stack components directly from the dashboard, along with improvements to service connectors, model artifact handling, and documentation. This release streamlines ML workflows with better component management capabilities, enhanced SageMaker integration, and critical fixes for custom flavor components and sorting logic.
Learn how to migrate from cnvrg.io to ZenML's open-source MLOps framework. Discover a sustainable alternative before Intel Tiber AI Studio's 2025 end-of-life. Get started with your MLOps transition today.
Learn how to build, fine-tune, and deploy multimodal LLMs using ZenML. Explore LLMOps best practices for deployment, real-time inference and model management.
ZenML 0.74.0 introduces key cloud provider features including SageMaker pipeline scheduling, Azure Container Registry implicit authentication, and Vertex AI persistent resource support. The release adds API Tokens for secure, time-boxed API authentication while delivering comprehensive improvements to timezone handling, database performance, and Helm chart deployments.
Our monthly roundup: AI Infrastructure Summit insights, new experiment comparison tools, and a deep dive into AI Engineering roles
Discover how organizations can transform their machine learning operations from manual, time-consuming processes into streamlined, automated workflows. This comprehensive guide explores common challenges in scaling MLOps, including infrastructure management, model deployment, and monitoring across different modalities. Learn practical strategies for implementing reproducible workflows, infrastructure abstraction, and comprehensive observability while maintaining security and compliance. Whether you're dealing with growing pains in ML operations or planning for future scale, this article provides actionable insights for building a robust, future-proof MLOps foundation.
ZenML 0.70.0 has launched with major improvements but requires careful handling during upgrade due to significant database schema changes. Key highlights include enhanced artifact versioning with batch processing capabilities, improved scalability through reduced server requests, unified metadata management via the new log_metadata method, and flexible filtering with the new oneof operator. The release also features expanded documentation covering finetuning and LLM/ML engineering resources. Due to the database changes, users must back up their data and test the upgrade in a non-production environment before deploying to production systems.
Why use ZenML alongside AWS / GCP / Azure MLOps platforms? Let's dive into why ZenML complements and enhance existing cloud MLOps infrastructure.
The combination of ZenML and Neptune can streamline machine learning workflows and provide unprecedented visibility into experiments. ZenML is an extensible framework for creating production-ready pipelines, while Neptune is a metadata store for MLOps. When combined, these tools offer a robust solution for managing the entire ML lifecycle, from experimentation to production. The combination of these tools can significantly accelerate the development process, especially when working with complex tasks like language model fine-tuning. This integration offers the ability to focus more on innovating and less on managing the intricacies of your ML pipelines.
This blog post discusses the integration of ZenML and BentoML in machine learning workflows, highlighting their synergy that simplifies and streamlines model deployment. ZenML is an open-source MLOps framework designed to create portable, production-ready pipelines, while BentoML is an open-source framework for machine learning model serving. When combined, these tools allow data scientists and ML engineers to streamline their workflows, focusing on building better models rather than managing deployment infrastructure. The combination offers several advantages, including simplified model packaging, local and container-based deployment, automatic versioning and tracking, cloud readiness, standardized deployment workflow, and framework-agnostic serving.
The combination of ZenML and Neptune can streamline machine learning workflows and provide unprecedented visibility into experiments. ZenML is an extensible framework for creating production-ready pipelines, while Neptune is a metadata store for MLOps. When combined, these tools offer a robust solution for managing the entire ML lifecycle, from experimentation to production. The combination of these tools can significantly accelerate the development process, especially when working with complex tasks like language model fine-tuning. This integration offers the ability to focus more on innovating and less on managing the intricacies of your ML pipelines.
The integration of ZenML and Databricks streamlines LLM development and deployment processes, offering scalability, reproducibility, efficiency, collaboration, and monitoring capabilities. This approach enables data scientists and ML engineers to focus on innovation.
This blog post discusses the integration of ZenML and Comet, an open-source machine learning pipeline management platform, to enhance the experimentation process. ZenML is an extensible framework for creating portable, production-ready pipelines, while Comet is a platform for tracking, comparing, explaining, and optimizing experiments and models. The combination offers seamless experiment tracking, enhanced visibility, simplified workflow, improved collaboration, and flexible configuration. The process involves installing ZenML and enabling Comet integration, registering the Comet experiment tracker in the ZenML stack, and customizing experiment settings.
Machine Learning Operations (MLOps) is crucial in today's tech landscape, even with the rise of Large Language Models (LLMs). Implementing MLOps on AWS, leveraging services like SageMaker, ECR, S3, EC2, and EKS, can enhance productivity and streamline workflows. ZenML, an open-source MLOps framework, simplifies the integration and management of these services, enabling seamless transitions between AWS components. MLOps pipelines consist of Orchestrators, Artifact Stores, Container Registry, Model Deployers, and Step Operators. AWS offers a suite of managed services, such as ECR, S3, and EC2, but careful planning and configuration are required for a cohesive MLOps workflow.
The ZenML team has addressed a security finding in ZenML Pro's role management system, reported by JFrog Security Research team. This update provides important information for users regarding role-based access controls and recommended actions
A new ZenML newsletter featuring Istanbul cooking adventures, faster docker builds, and more
The combination of ZenML and SkyPilot offers a robust solution for managing ML workflows.
Master cloud-based LLM finetuning: Set up infrastructure, run pipelines, and manage experiments with ZenML's Model Control Plane for Microsoft's latest Phi model.
Playing around with some genAI services and tools to create a story and comic that showcases the journey of MLOps adoption for a small team.
Master cloud-based LLM finetuning: Set up infrastructure, run pipelines, and manage experiments with ZenML's Model Control Plane for Meta's latest Llama model.
Learn how to leverage caching, parameterization, and smart infrastructure switching to iterate faster on machine learning projects while maintaining reproducibility.
Shipping 🤗 datasets visualization embedded in the ZenML dashboard in a few hours
Streamline your machine learning platform with ZenML. Learn how ZenML's 1-click cloud stack deployments simplify setting up MLOps pipelines on AWS, GCP, and Azure.
On the difficulties in precisely defining a machine learning pipeline, exploring how code changes, versioning, and naming conventions complicate the concept in MLOps frameworks like ZenML.
We dive deep into the world of Retrieval-Augmented Generation (RAG) pipelines and how ZenML can streamline your RAG workflows.
Today, we're back to LLM land (Not too far from Lalaland). Not only do we have a new LoRA + Accelerate-powered finetuning pipeline for you, we're also hosting a RAG themed webinar.
We've open-sourced our new dashboard to unify the experience for OSS and cloud users, although some features are initially CLI-only. This launch enhances onboarding and simplifies maintenance. Cloud users will see no change, while OSS users can enjoy a new interface and DAG visualizer. We encourage community contributions to help us expand and refine this dashboard further, looking forward to integrating more features soon.
Community member Marwan Zaarab explains how and why he built a VS Code Extension for ZenML.
A critical security vulnerability has been identified in ZenML versions prior to 0.46.7. This vulnerability potentially allows unauthorized users to take ownership of ZenML accounts through the user activation feature.
ZenML secures an additional $3.7M in funding led by Point Nine, bringing its total Seed Round to $6.4M, to further its mission of simplifying MLOps. The startup is set to launch ZenML Cloud, a managed service with advanced features, while continuing to expand its open-source framework.
We released an updated way to deploy MLOps infrastructure, building on the success of the `mlops-stack` repo and its stack recipes. All the new goodies are available via the `mlstacks` Python package.
The 0.40.0 release introduces a completely reworked interface for developing your ZenML steps and pipelines. It makes working with these components much more natural, intuitive, and enjoyable.
Explore how ZenML, an MLOps framework, can be used with large language models (LLMs) like GPT-4 to analyze and version data from databases like Supabase. In this case study, we examine the you-tldr.com website, showcasing ZenML pipelines asynchronously processing video data and generating summaries with GPT-4. Understand how to tackle large language model limitations by versioning data and comparing summaries to unlock your data's potential. Learn how this approach can be easily adapted to work with other databases and LLMs, providing flexibility and versatility for your specific needs.
We decided to explore how the emerging technologies around Large Language Models (LLMs) could seamlessly fit into ZenML's MLOps workflows and standards. We created and deployed a Slack bot to provide community support.
ZenNews is a tool powered by ZenML that can automate the summarization of news sources and save you time and effort while providing you with the information you need.
Getting started with your ML project work is easier than ever with Project Templates, a new way to generate scaffolding and a skeleton project structure based on best practices.
A winning entry - 2nd prize winner at Month of MLOps 2022 competition.
ZenML 0.23.0 comes with a brand-new experiment tracker flavor - Neptune.ai! We dive deeper in this blog post.
The ZenML MLOps Competition ran from October 10 to November 11, 2022, and was a wonderful expression of open-source MLOps problem-solving.
Transform quickstart PyTorch code as a ZenML pipeline and add experiment tracking and secrets manager component.
Test automation is tedious enough with traditional software engineering, but machine learning complexities can make it even less appealing. Using Deepchecks with ZenML pipelines can get you started as quickly as it takes you to read this article.
How to use ZenML and KServe to deploy serverless ML models in just a few steps.
ZenML combines forces with Great Expectations to add data validation to the list of continuous processes automated with MLOps. Discover why data validation is an important part of MLOps and try the new integration with a hands-on tutorial.
Getting started with distributed ML in the cloud: How to orchestrate ML workflows natively on Amazon Elastic Kubernetes Service (EKS).
How ZenML lets you have the best of both worlds, serverless managed infrastructure without the vendor lock in.
As our AI/ML projects evolve and mature, our processes and tooling also need to keep up with the growing demand for automation, quality and performance. But how can we possibly reconcile our need for flexibility with the overwhelming complexity of a continuously evolving ecosystem of tools and technologies? MLOps frameworks promise to deliver the ideal balance between flexibility, usability and maintainability, but not all MLOps frameworks are created equal. In this post, I take a critical look at what makes an MLOps framework worth using and what you should expect from one.
With ZenML 0.6.3, you can now run your ZenML steps on Sagemaker, Vertex AI, and AzureML! It’s normal to have certain steps that require specific infrastructure (e.g. a GPU-enabled environment) on which to run model training, and Step Operators give you the power to switch out infrastructure for individual steps to support this.
An exploration of some frameworks created by Google and Microsoft that can help think through improvements to how machine learning models get developed and deployed in production.
This week I spoke with Johnny Greco, a data scientist working at Radiology Partners. Johnny transitioned into his current work from a career as an academic — working in astronomy — where also worked in the open-source space to build a really interesting synthetic image data project.
Connecting model training pipelines to deploying models in production is regarded as a difficult milestone on the way to achieving Machine Learning operations maturity for an organization. ZenML rises to the challenge and introduces a novel approach to continuous model deployment that renders a smooth transition from experimentation to production.
We recently reworked a number of parts of our CLI interface. Here are some quick wins we implemented along the way that can help you improve how users interact with your CLI via the popular open-source library, rich.
Tristan and Alex discuss where machine learning and AI are headed in terms of the tooling landscape. Tristan outlined a vision of a higher abstraction level, something he's working on making a reality as CEO at Continual.
Use MLflow Tracking to automatically ensure that you're capturing data, metadata and hyperparameters that contribute to how you are training your models. Use the UI interface to compare experiments, and let ZenML handle the boring setup details.
Mohan and Alex discuss neurosymbolic AI and the implications of a shift towards that as a core paradigm for production AI systems. In particular, we discuss the practical consequences of such a shift, both in terms of team composition as well as infrastructure requirements.
We discuss how to monitor models in production, and how it helps you in the long-run.
All the advantages that ZenML will bring you if you choose to use it to productionize your model development workflows.
Use caches to save time in your training cycles, and potentially to save some money as well!
We discuss the role of MLOps in an organization, some deployment war stories from his career as well as what he considers to be 'best practices' in production machine learning.
We launched a podcast to have conversations with people working to productionize their machine learning models and to learn from their experience.
Why data scientists need to own their ML workflows in production.
Eliminate technical debt with iterative, reproducible pipelines.
Short answer: not really, but it can become better!
An overview of some of the capabilities that ZenML will unlock for our users.
A set of guiding principles to help you better productionize your machine learning models.
Pipelines help you think and act better when it comes to how you execute your machine learning training workflows.
ZenML makes it easy to setup training pipelines that give you all the benefits of cached steps.
