Deploying AI agents in enterprises is complex, balancing security, scalability, and usability. This post compares deployment paths on Google Cloud—highlighting Cloud Run with IAP as the most secure and flexible option—and shows how teams can build powerful agents with ADK without losing the human touch.

Deploying AI Agents in the Enterprise without Losing your Humanity using ADK and Google Cloud

How do fine-tuned models and RAG systems power personalized AI agents that learn, collaborate, and transform enterprise workflows? What kind of technical challenges do we need to first examine before this becomes real?

Fine-Tuned Models Are Getting Out of Hand

As AI agents become more capable, their real-world performance increasingly depends on how well they can coordinate tools.

This month's paper introduces a benchmark designed to rigorously test how AI agents handle multi-step tasks using the Model Context Protocol (MCP) — the emerging standard for tool integration.

​The authors present 101 carefully curated real-world queries, refined through iterative LLM rewriting and human review, that challenge models to coordinate multiple tools such as web search, file operations, mathematical reasoning, and data analysis.

Stress Testing and Diagnosing MCP-enabled Agents on Challenging Queries

Modern LLMs are defined as much by how they’re trained as by what they learn. This post unpacks the often-overlooked foundations of that process: pretraining—the stage that shapes a model’s core reasoning and knowledge. Starting with ULMFiT’s breakthrough in transfer learning and InstructGPT’s formalized multi-stage pipeline, it explores how pretraining has evolved into a dynamic ecosystem of techniques, from instruction-augmented and multi-phase approaches to continual and reinforcement-based pretraining. Amid the growing complexity and shifting definitions, one truth remains: understanding pretraining is essential to understanding how language models think, reason, and behave.

Pretraining: Breaking Down the Modern LLM Training Pipeline

<h1>The Virtual AI Event That’s Actually... Fun</h1>We know what you’re thinking: another virtual conference. Another day of talking heads, awkward silences, and the constant urge to check your email. You’ve been there. You get distracted, you drop off, and you end up missing the one talk you needed. The truth is, most virtual events are a snooze-fest.That's not how we do it.<h3>100% AI Agents in Prod. BUT This Isn't Just Another Zoom Link.</h3>This is the 6th Annual Virtual Conference. We’re not just packing the day with the biggest minds building AI agents; we’re making sure you stay hooked from the first moment to the last.Last year, we saw folks start talking about agents. Companies stopped playing with them and started building real-world solutions. We got the best around to come talk about the challenges they had to overcome to get the new technology working.That's why we're doubling down on the progress, lessons, and breakthroughs that are truly shaping the future. Our theme is once again Agents in Production. You'll get cutting-edge, actionable insights that will move you from experimentation to legit deployment.<h2>30+ Talks on AI Agents. I Promise You Won’t Log Off Early</h2>We’ve engineered this high-energy virtual event to be FUN. Sitting still for hours is impossible, so we're building in the kind of spontaneous, engaging moments that make live events great:<ol><li data-list="bullet">Hilarious Skits: Short, professionally-produced, and funny skits woven between sessions to break up the talks and give your brain a laugh.</li><li data-list="bullet">Improvised Live Music: Our in-house musicians will spontaneously compose songs based on the session topics, turning a technical deep-dive into an unforgettable performance.</li><li data-list="bullet">High-Energy Engagement: You’ll be treated like a live audience member, not a muted participant.</li></ol>The most important piece of the event is the deep technical content that our presenters bring. But in the breaks, we make a fool of ourselves. It's a conference for your brain AND your attention span.<h2>What You'll Miss If You Stay Home</h2>If you're not with us on November 18th, you’ll be missing out on:<ol><li data-list="bullet">Real Stories from Real Practitioners</li><li data-list="bullet">The Hard-Won Lessons: Inspiring use cases and practical applications that tell you exactly how the leading companies are successfully deploying agentic AI.</li><li data-list="bullet">The Deep Dives: Hands-on technical workshops and talks from the voices defining the future of AI.</li><li data-list="bullet">The Connections: Global opportunities to collaborate, share, and grow within the ML community.</li></ol>This is your chance to get up to speed on the global AI scene, connect with innovators, and actually enjoy a virtual event.

Agents in Production - MLOps x Prosus

AI Agent Builder Summit SF

<h1>AI Agents Are Already Working — Let’s Talk About It</h1>Agents are no longer just experiments. From e-commerce to customer support to analytics workflows, they’re quietly getting real work done in production.On July 17, join the MLOps Community for Part 2 of Agents in Production — a virtual event focused on the messy, practical side of building and deploying AI agents.What’s on deck?<ol><li data-list="bullet">Taming complexity: agent memory, behavior control, latency vs. response tradeoffs</li><li data-list="bullet">Stories from the field: How companies are actually using agents in live environments</li><li data-list="bullet">Tooling that works: routing, evaluation, UX, and cost performance in the wild</li></ol><h2>It’s free, it’s global, and it’s going to be packed. </h2>

Agents in Production 2025

Your AI shouldn’t just work in the demo. It should work in production.Join us for a high-impact, no-fluff Mini Summit focused on building AI applications that can withstand the real world — where APIs fail, users take days to respond, and servers crash without warning. Whether you’re deep into building with LLMs or just starting to explore how to operationalize your AI systems, this event is packed with practical insights, real-world strategies, and modern tooling to help you ship reliable AI that doesn’t break. What’s on the agenda?🔹 Process Calling: Agentic Tools Need State Function calling gave LLMs a way to "do" things — but it’s not enough. When you’re building agents for customer-facing use cases, stateless abstractions fall short fast. Learn why the future of agentic tooling is process-based, not function-based, and what it means to build agents that remember, recover, and reliably finish what they start. 🔹 Building Reliable AI Applications with Durable Workflows Chaining functions together is easy. Keeping AI workflows running when things go sideways? That’s the hard part. This talk introduces durable workflows — systems that checkpoint state, recover automatically, and gracefully handle everything from human delays to API flakiness. You’ll see real examples of AI pipelines that stay resilient in production. 🔹 No YAML? No Problem: Orchestrate Kubernetes Workflows the Easy Way with Python Sick of writing orchestration logic in YAML? You’re not alone. Discover how Hera, the Python SDK for Argo Workflows, lets you express complex Kubernetes workflows using clean, testable Python code. Keep your business logic and orchestration logic in one place — no indentation nightmares required. Who should attend? AI engineers, MLOps professionals, infra leads, and builders who care about more than just flashy demos. If you're looking to make your AI actually work at scale, this one's for you. RSVP now and start building AI that doesn’t break.

Building AI That Doesn’t Break

<h1>🔥✨ GenAI in Games, 3D Animation &amp; VFX ✨🔥</h1><h3>🗓 June 10th @ 10:00 -11:30 AM PST Pacific West-Coast Time 19:00 - 20:30 CEST Central Eastern Time</h3>🔥Hot Topics 🔥:<ol><li data-list="bullet">AI in Game Development: Friction or Fruition?</li><li data-list="bullet">Demo GenAI game tools usage in pre-production</li><li data-list="bullet">Beyond the Prompt: Using AI as a tool to accelerate execution, not replace creative intent</li><li data-list="bullet">Legal Playbook for AI Creativity</li></ol>Calling all creative visionaries—producers, PMs, creative directors, and pipeline innovators. Join us for a future-forward session exploring how Generative AI is transforming entertainment studios—from previs to post-production. Whether you're navigating massive asset libraries, complex feedback loops, or tightening delivery pipelines, this is your chance to rethink workflows from the ground up—technically, creatively, and legally.Hear from a diverse lineup of presenters, each bringing their unique perspective—from real-time GenAI integration and creative collaboration tools to pipeline automation, IP and legal frameworks, and AI-assisted worldbuilding.Discover how studios are using GenAI to unlock smarter systems, efficient iteration, and more reflective space for creativity, all while addressing the evolving ethical and legal challenges shaping the future of our industry.If you're crafting cinematic universes or building next-gen experiences at scale, this is where innovation meets responsible storytelling power.

GenAi in Games, 3D Animation and VFX

Home

Forum

Events

Content

People

Contributors

Albums

Clubs

Chat

Channels

Tools

Help

The MLOps Community is where machine learning practitioners come together to define and implement MLOps.
Our global community is the default hub for MLOps practitioners to meet other MLOps industry professionals, share their real-world experience and challenges, learn skills and best practices, and collaborate on projects and employment opportunities. We are the world's largest community dedicated to addressing the unique technical and operational challenges of production machine learning systems.

MLOps Community

Job Title

Login to the MLOps Community community to connect with others, attend community events, and more!

Login or Sign Up for even more

To unlock all parts of the community and get the best experience, complete your profile.