{ "version": "https://jsonfeed.org/version/1.1", "title": "Valohai Blog", "home_page_url": "https://valohai.com/blog/", "feed_url": "https://valohai.com/blog/feed.json", "items": [ { "id": "https://valohai.com/blog/which-llm-should-power-your-support-bot/", "url": "https://valohai.com/blog/which-llm-should-power-your-support-bot/", "title": "Which LLM Should Power Your Support Bot? How Systematic Evaluation Turned a Gut Feeling Into a Data-Backed Decision", "content_text": "When you're building a product on top of an LLM, there's a moment everyone hits eventually. You've got a working prototype, it feels pretty good, and now you need to decide which model to ship with. Here's what a systematic evaluation approach looks like, and what it can reveal.", "date_published": "2026-03-03T09:00:00.000Z", "date_modified": "2026-03-03T09:00:00.000Z", "image": "https://valohai.com/_astro/LLMEvaluation.CtKqN0gb.webp", "authors": [ { "name": "Petteri Raatikainen", "url": "https://valohai.com/blog/author/petteri-raatikainen/" } ], "tags": [ "AI", "MLOps" ] }, { "id": "https://valohai.com/blog/ai-factory-series-waiting/", "url": "https://valohai.com/blog/ai-factory-series-waiting/", "title": "Half Your Team Is Waiting for the Other Half (And It's Killing You)", "content_text": "Your smartest people are stuck waiting. Not because they're lazy, but because your systems waste their time. This is Part 1 of the AI Factory Series on how Lean principles apply to scaling AI teams.", "date_published": "2025-12-15T09:00:00.000Z", "date_modified": "2025-12-15T09:00:00.000Z", "image": "https://valohai.com/_astro/ai-factory-series-waiting.DLQaEsB6.png", "authors": [ { "name": "Rasmus Bentsen", "url": "https://valohai.com/blog/author/rasmus-bentsen/" } ], "tags": [ "AI Factory" ] }, { "id": "https://valohai.com/blog/continental-predictive-ml-production/", "url": "https://valohai.com/blog/continental-predictive-ml-production/", "title": "Deploying Continental R&D’s First Predictive ML Model: How an Industrial Giant Scaled Multilingual Machine Learning Into Production", "content_text": "How Continental Tires deployed their first predictive ML model into production, reduced testing cycles from two months to overnight, and built a multilingual R-and-Python pipeline that survives real-world industrial complexity.", "date_published": "2025-12-02T12:00:12.000Z", "date_modified": "2025-12-02T00:00:00.000Z", "image": "https://valohai.com/_astro/continental-blog.DiR9j5EL.webp", "authors": [ { "name": "Toni Perämäki", "url": "https://valohai.com/blog/author/toni-per%C3%A4m%C3%A4ki/" } ], "tags": [ "MLOps", "Manufacturing", "Case Study" ] }, { "id": "https://valohai.com/blog/why-most-alphafold-pipelines-fail/", "url": "https://valohai.com/blog/why-most-alphafold-pipelines-fail/", "title": "Why Most AlphaFold Pipelines Fail at Scale (And What to Do Instead)", "content_text": "Your AlphaFold pipeline is probably held together with bash scripts and hope. Here's why that's not your fault and how to fix it.", "date_published": "2025-09-24T00:00:00.000Z", "date_modified": "2025-09-24T00:00:00.000Z", "image": "https://valohai.com/_astro/alphafold.29q9vBIE.png", "authors": [ { "name": "Drazen Dodik", "url": "https://valohai.com/blog/author/drazen-dodik/" } ], "tags": [ "MLOps" ] }, { "id": "https://valohai.com/blog/valohai-on-oracle-cloud-sagemaker-like-mlops/", "url": "https://valohai.com/blog/valohai-on-oracle-cloud-sagemaker-like-mlops/", "title": "Valohai’s MLOps Platform Now Available on Oracle Cloud Marketplace", "content_text": "Valohai’s enterprise-grade MLOps platform is now available on Oracle Cloud Marketplace, bringing SageMaker-like MLOps on OCI with automated pipelines, governance, and scalability for regulated industries.", "date_published": "2025-09-23T14:00:00.000Z", "date_modified": "2025-09-23T14:00:00.000Z", "image": "https://valohai.com/_astro/valohai-oracle-partnership.rPXHer-F.png", "authors": [ { "name": "Toni Perämäki", "url": "https://valohai.com/blog/author/toni-per%C3%A4m%C3%A4ki/" } ], "tags": [ "Valohai", "MLOps" ] }, { "id": "https://valohai.com/blog/productivity-dashboard/", "url": "https://valohai.com/blog/productivity-dashboard/", "title": "See the Bigger Picture: Valohai's Productivity Dashboard Delivers Complete ML Operations Visibility", "content_text": "Valohai's new Productivity Dashboard visualizes ML operations ROI, offering insights into cost savings, innovation acceleration, operational excellence, and governance for stakeholders.", "date_published": "2025-08-13T12:15:24.807Z", "date_modified": "2025-08-13T12:24:45.087Z", "image": "https://valohai.com/_astro/increasing-productivity.BkgdHwqy.png", "authors": [ { "name": "Petteri Raatikainen", "url": "https://valohai.com/blog/author/petteri-raatikainen/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/mlflow-vs-enterprise-mlops/", "url": "https://valohai.com/blog/mlflow-vs-enterprise-mlops/", "title": "MLflow vs Enterprise MLOps: When to Switch from Open Source to Platform", "content_text": "MLflow democratized ML experiment tracking, but features that mature teams need feel like afterthoughts. When growing teams discover that what worked for 5 people breaks at 50, it might be time to consider a platform approach.", "date_published": "2025-07-16T09:00:00.000Z", "date_modified": "2025-07-16T09:00:00.000Z", "image": "https://valohai.com/_astro/mlflow-vs-enterprise-mlops.gcqV-p3E.png", "authors": [ { "name": "Aicha Moussaid" } ], "tags": [ "Valohai", "MLOps", "MLflow" ] }, { "id": "https://valohai.com/blog/how-valohai-supercharges-nvidia-monai/", "url": "https://valohai.com/blog/how-valohai-supercharges-nvidia-monai/", "title": "Scaling Medical Imaging AI with Confidence: How Valohai Supercharges NVIDIA MONAI", "content_text": "NVIDIA MONAI powers cutting-edge medical imaging research. Here’s how Valohai makes it reproducible, scalable, and production-ready — even in highly regulated healthcare environments.", "date_published": "2025-06-11T09:00:00.000Z", "date_modified": "2025-06-11T09:00:00.000Z", "image": "https://valohai.com/_astro/nvidia-monai-valohai.QcNrc81m.png", "authors": [ { "name": "Toni Perämäki", "url": "https://valohai.com/blog/author/toni-per%C3%A4m%C3%A4ki/" } ], "tags": [ "Valohai", "MLOps" ] }, { "id": "https://valohai.com/blog/how-valohai-supercharges-nvidia-nemo/", "url": "https://valohai.com/blog/how-valohai-supercharges-nvidia-nemo/", "title": "Scaling Speech AI with Ease: How Valohai Supercharges NVIDIA NeMo", "content_text": "NVIDIA NeMo brings state-of-the-art speech AI to the enterprise. Here’s how Valohai makes it production-ready without leaving your ML team with hand-stitched scripts, notebooks, and wishful thinking.", "date_published": "2025-06-06T09:00:00.000Z", "date_modified": "2025-06-06T09:00:00.000Z", "image": "https://valohai.com/_astro/nvidia-nemo-valohai.Devr0Wz2.png", "authors": [ { "name": "Toni Perämäki", "url": "https://valohai.com/blog/author/toni-per%C3%A4m%C3%A4ki/" } ], "tags": [ "Valohai", "MLOps" ] }, { "id": "https://valohai.com/blog/sagemaker-true-cost/", "url": "https://valohai.com/blog/sagemaker-true-cost/", "title": "Stop Making Your Data Scientists Learn AWS: The True Cost of SageMaker", "content_text": "AWS SageMaker promises end-to-end ML workflows, but the hidden complexity often turns your data scientists into part-time DevOps engineers. Here's what that really costs your team.", "date_published": "2025-05-21T09:00:00.000Z", "date_modified": "2025-05-21T09:00:00.000Z", "image": "https://valohai.com/_astro/sagemaker-true-cost.qwSBv7uR.png", "authors": [ { "name": "Drazen Dodik", "url": "https://valohai.com/blog/author/drazen-dodik/" } ], "tags": [ "Valohai", "MLOps" ] }, { "id": "https://valohai.com/blog/mlops-platforms-code-coupling/", "url": "https://valohai.com/blog/mlops-platforms-code-coupling/", "title": "Why Most MLOps Platforms Want a Deep Relationship with Your Codebase (and Why We Don't)", "content_text": "Most MLOps platforms quietly couple themselves to your codebase—until one day you're debugging their wrappers instead of your model. Here's why that happens, what it costs you, and how to avoid it entirely.", "date_published": "2025-05-12T10:00:00.000Z", "date_modified": "2025-05-12T10:00:00.000Z", "image": "https://valohai.com/_astro/mlops-platforms-code-coupling.Dvnl0TLE.png", "authors": [ { "name": "Drazen Dodik", "url": "https://valohai.com/blog/author/drazen-dodik/" } ], "tags": [ "MLOps", "Code Coupling", "SDK", "Valohai", "API" ] }, { "id": "https://valohai.com/blog/ml-reproducibility-crisis/", "url": "https://valohai.com/blog/ml-reproducibility-crisis/", "title": "The Hidden Reproducibility Crisis Killing Your ML Team's Productivity (And Your Budget)", "content_text": "ML reproducibility challenges rob teams of productivity and inflate costs through duplicate experiments. This post explores how automated reproducibility can save compute costs, accelerate development, and eliminate the frustration of recreating successful models when they matter most.", "date_published": "2025-04-30T10:00:00.000Z", "date_modified": "2025-04-30T10:00:00.000Z", "image": "https://valohai.com/_astro/ml-reproducibility-crisis.f4ayji7i.png", "authors": [ { "name": "Drazen Dodik", "url": "https://valohai.com/blog/author/drazen-dodik/" } ], "tags": [ "MLOps", "ML Reproducibility", "Experiment Tracking", "Valohai" ] }, { "id": "https://valohai.com/blog/the-mlflow-airflow-kubernets-makeshift-monster/", "url": "https://valohai.com/blog/the-mlflow-airflow-kubernets-makeshift-monster/", "title": "The MLFlow-Airflow-Kubernetes Makeshift Monster: How Your DIY ML Stack Became Your Biggest Bottleneck", "content_text": "Handling ML infrastructure with MLFlow, Airflow, and Kubernetes often creates a monster that devours team productivity. Learn how knowledge silos form, why debugging becomes a nightmare, and how to escape your ML stack prison without burning everything down or disrupting your team's workflow.", "date_published": "2025-04-28T10:00:00.000Z", "date_modified": "2025-04-28T10:00:00.000Z", "image": "https://valohai.com/_astro/the-dyi-ml-stack-monster.DUpPNFEf.png", "authors": [ { "name": "Drazen Dodik", "url": "https://valohai.com/blog/author/drazen-dodik/" } ], "tags": [ "MLOps", "Valohai", "ML Infrastructure" ] }, { "id": "https://valohai.com/blog/how-to-manage-massive-datasets-in-valohai/", "url": "https://valohai.com/blog/how-to-manage-massive-datasets-in-valohai/", "title": "How to manage massive datasets in Valohai", "content_text": "Handling massive datasets can be time-consuming and error-prone. We are introducing multiple additions to Valohai designed to streamline ML workflows involving a massive number of files, from dataset creation and preprocessing to model training and data lineage tracking.", "date_published": "2025-02-19T09:00:00.000Z", "date_modified": "2025-02-19T09:00:00.000Z", "image": "https://valohai.com/_astro/how-to-manage-massive-datasets-in-valohai.oQcQ4SZO.webp", "authors": [ { "name": "Tomi Kokkonen", "url": "https://valohai.com/blog/author/tomi-kokkonen/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/2024-in-review-part-1/", "url": "https://valohai.com/blog/2024-in-review-part-1/", "title": "2024 in Review (Part 1)", "content_text": "Let's take a look back at the past year. In this first part of our annual review, we'll recap all the key additions and improvements to our end-to-end MLOps platform, ecosystem integrations, and more. Stick around and you'll find out what to expect in the next year and far beyond!", "date_published": "2024-12-18T09:00:00.000Z", "date_modified": "2024-12-18T09:00:00.000Z", "image": "https://valohai.com/_astro/2024-in-review-part-1.CpzEK2EX.png", "authors": [ { "name": "Tarek Oraby", "url": "https://valohai.com/blog/author/tarek-oraby/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/boosting-velocity-in-data-science-teams/", "url": "https://valohai.com/blog/boosting-velocity-in-data-science-teams/", "title": "Boosting Velocity in Data Science Teams: A Practical Guide", "content_text": "Create structured and efficient workflows that help your data science team work faster and smarter, i.e., maximize the impact on the business and increase the speed of experimentation and delivery without compromising quality.", "date_published": "2024-11-27T09:00:00.000Z", "date_modified": "2024-11-27T09:00:00.000Z", "image": "https://valohai.com/_astro/velocity-in-data-science.CbwHaJzd.png", "authors": [ { "name": "Tarek Oraby", "url": "https://valohai.com/blog/author/tarek-oraby/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/stop-wasting-your-gpus-with-valohais-dynamic-gpu-allocation/", "url": "https://valohai.com/blog/stop-wasting-your-gpus-with-valohais-dynamic-gpu-allocation/", "title": "Stop wasting your GPUs with Valohai's Dynamic GPU Allocation", "content_text": "Our latest feature is built to help you make the most out of your on-prem hardware: utilize idle GPUs, adjust GPU usage for every ML job, and forget about managing priority queues. It’s live and ready for you to give it a spin (no pun intended).", "date_published": "2024-11-20T09:00:00.000Z", "date_modified": "2024-11-20T09:00:00.000Z", "image": "https://valohai.com/_astro/dynamic-gpu-allocation.Bm05Stpp.png", "authors": [ { "name": "Tarek Oraby", "url": "https://valohai.com/blog/author/tarek-oraby/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/valohai-audit-log-traceability-built-for-ai-governance/", "url": "https://valohai.com/blog/valohai-audit-log-traceability-built-for-ai-governance/", "title": "Valohai's Audit Log: Traceability built for AI governance", "content_text": "Introducing an out-of-the-box solution that gives all Valohai users automatic, immutable, and secure audit logs that ensure traceability for navigating compliance requirements, debugging issues, and improving accountability within teams.", "date_published": "2024-11-06T09:00:00.000Z", "date_modified": "2024-11-06T09:00:00.000Z", "image": "https://valohai.com/_astro/audit-log.BsCm6xG7.png", "authors": [ { "name": "Tarek Oraby", "url": "https://valohai.com/blog/author/tarek-oraby/" } ], "tags": [ "Valohai" ] }, { "id": "https://valohai.com/blog/amd-gpu-performance-for-llm-inference/", "url": "https://valohai.com/blog/amd-gpu-performance-for-llm-inference/", "title": "AMD GPU Performance for LLM Inference: A Deep Dive", "content_text": "AMD's MI300X GPU can outperform Nvidia's H100 in LLM inference benchmarks, offering larger memory and higher bandwidth. Read our benchmark in full, get the details, and discover how this impacts AI hardware performance and model capabilities.", "date_published": "2024-10-31T13:52:56+03:00", "date_modified": "2024-10-31T13:52:56+03:00", "image": "https://valohai.com/_astro/gpu-performance-llm-inference.CM-e_tFq.png", "authors": [ { "name": "Eero Laaksonen", "url": "https://valohai.com/blog/author/eero-laaksonen/" } ], "tags": [ "MLOps" ] }, { "id": "https://valohai.com/blog/simplify-and-automate-the-machine-learning-model-lifecycle/", "url": "https://valohai.com/blog/simplify-and-automate-the-machine-learning-model-lifecycle/", "title": "Simplify and automate the machine learning model lifecycle", "content_text": "We’ve built the Model Hub to help you streamline and automate model lifecycle management. Leverage Valohai for lineage tracking, performance comparison, workflow automation, access control, regulatory compliance, and more.", "date_published": "2024-09-18T09:00:00.000Z", "date_modified": "2024-09-18T09:00:00.000Z", "image": "https://valohai.com/_astro/model-hub.BArxP1B4.png", "authors": [ { "name": "Tarek Oraby", "url": "https://valohai.com/blog/author/tarek-oraby/" } ], "tags": [ "Valohai" ] } ] }