Union.ai

This workshop will equip you with the skills to effectively build your computer vision and ML pipelines using Python, PyTorch and Union. Links to follow along live: - Sign up for Union: https://www.union.ai/ - Open the repo in colab: https://lnkd.in/gZ6-mprs We'll cover data annotation for object detection, model fine-tuning, versioning, building an efficient AI pipeline, and running inference on the trained CV model locally. Afterwards you'll be able to build your own computer vision object detection models and custom datasets! What you'll learn can also be transferred to different types of AI and ML pipelines.

Attention #ai #ml and #datascience pros - Union’s December Newsletter just dropped! 🔥🗞️ Learn about new features, upcoming events, and job openings. By the way, you can also schedule a FREE consultation with someone from our team of expert AI/ML engineers (link in newsletter) Read the full post for details and get more info on how to schedule a free 1:1 consultation with an AI/ML engineer.

💲 👀 Check out the latest from our Unified AI Platform - Cost Observability. Learn how a lack of transparency into your expenses is dragging down your bottom line, and how Union's Cost Allocation Dashboard puts your money - and development time - back in your pocket. https://hubs.la/Q02_h2-Q0

We're going live at 11:00am PT to talk about secure AI orchestration and how to mitigate model-centric attacks! Major progress in machine learning (ML) has led to a corresponding boom in the broader artificial intelligence (AI) space, opening up commercial applications in text, image, audio, and video generation. However, data scientists and ML engineers still face many security issues that may lead to arbitrary code execution even in the space of "classical" ML, which often involves classification or regression on tabular data. This workshop will outline and prepare you for two types of model-centric attacks: - Malicious code injection on pickled model files - Malicious code written and executed by an LLM Links to follow along: Union Account: https://www.union.ai/ Code: https://lnkd.in/gpDpT5rH

Thank you for having us in your #reinvent pavilion NVIDIA! We had a great time showcasing our integrations and how we enable better AI workflows 🤖

🛠️ Join our next workshop with Niels Bantilan on Secure AI Orchestration: Mitigate Model-centric Attacks 🔐 Major progress in machine learning (ML) has led to a corresponding boom in the broader artificial intelligence (AI) space, opening up commercial applications in text, image, audio, and video generation. However, data scientists and ML engineers still face many security issues that may lead to arbitrary code execution even in the space of "classical" ML, which often involves classification or regression on tabular data. This workshop will outline and prepare you for two types of model-centric attacks: - Malicious code injection on pickled model files - Malicious code written and executed by an LLM 🔗 https://lnkd.in/gfWGAtXC

4/4 Unification is in the name As we conclude the series of announcements from re:Invent (and near the end of year for a recap) I want to share, from my perspective, how we, as a company, got here and where we are heading... We started Union.ai 4 years ago with a goal of bringing different AI Systems together to make it cheap, easy, fast, efficient and optimal to build ML Systems. We built on the solid foundation of our open source project Flyte. Today, this vision is closer to reality than ever. In such a fast and ever evolving landscape of tools and paradigms, it's very hard and often impractical to build standards. In absence of standards, tight partnerships fill in the gaps. And as patterns emerge, standards will follow. Union partners with companies (startups as well as established players) serving various stages of the ML Lifecycle to bring a cohesive experience to the humans actually developing the systems. Today Union is an established partner with Amazon Web Services (AWS), NVIDIA (DGX, NIM and other technologies), Weights & Biases, neptune.ai, WhyLabs and many others. The world of AI may seem very disorienting and we aim at bringing it together so you can focus on building awesome stuff! Our journey is still in its infancy, we have quite a few things being baked and I can't wait to share more. Ping me if you are tired of all the acronyms you have to learn and all the benchmarks you have to sift through and all the nuances that may only become relevant when it's too late. Chances are we have an answer for you that fits what YOU do.

We had a great time on the ground at this year’s AWS re:Invent! We showed off how easy it is to build production-grade, scalable AI products with our Unified AI Platform. We also introduced Actors, our reusable containers that dramatically reduce cold-start time for workflows. Learn more about what we announced: https://lnkd.in/gzSWbQGe

⚙ In modern ML workflows, efficient data handling is essential for maximizing GPU utilization and accelerating training. ➡ Join us for the next community presentation where Shuying Liang will introduce the Flyte K8s agent LinkedIn Engineering is developing to orchestrate services that offload dataset loading and transformation tasks from individual training pipelines. This setup supports consistent, reusable data transformations across runs and dynamically scales to meet varying training workload demands. Built on the open-source Flyte agent framework, the Flyte K8s agent enables a decoupled, modular system architecture, allowing rapid development iterations and seamless transitions to production. At LinkedIn, this agent serves as a cornerstone in supporting scalable, adaptable data services across various training workflows, with Graph Neural Networks (GNN) as a prominent use case. This presentation delves into how the Flyte K8s agent’s general design drives productivity and operational excellence, making it an ideal solution for LinkedIn’s deep learning needs such as GNN. 💜 Everyone is welcome to join, learn, and ask questions!

Dealing with disconnected workflows, runaway infrastructure costs, or slow time-to-market for AI solutions? We’re at AWS re:Invent this week, demoing our Unified AI Platform that unlocks 25x faster dev cycles. Visit Union at the NVIDIA booth (# 1620) in the re:Invent Expo Hall. Can’t make it? Learn more here: https://lnkd.in/gqciBFtP