AI Development & MLOps
Practical aspects of developing, deploying, and managing AI and machine learning models in production, including MLOps practices, LLM operations, open-source tools like LangChain, and development environments like Replit.
Viral Moments
28
Fixing audio with words
Non-deterministic agents
Prototype vs. Production
AI as your co-pilot
AI is changing everything
Automated remediation
Build apps in minutes!
Always need evals
AI changes developer tools
Instant AI app deployment
Inferred errors explained
Reliability for production
Perfection takes time
Anyone can build now
Beyond single prompts
30th to 5th
Play with keynote code
Code, learn, iterate
AI building Google's products!
Optimize tokens, save capacity
Models need fresh data
The Inference Bible
Simplified evaluation process
Free AI learning!
Building with our own tools
Virtual file systems
ADK's powerful controls
Evolution of rapid dev
Key Quotes
11
“now your developers don't have to switch between Dart and JavaScript for the back end or Dart and Go. They can just stay in Dart.”
“We are moving from demo to something that is production that can run in production reliably and can really solve real problems.”
“It's very very difficult to turn that into something you can put in production reliably.”
“I like to have fairly long conversation about my idea, the tech stack, and then ask it to do the research if maybe, you know, my idea is not perfect or try to optimize it.”
“You had binaries before, now you have skills. As dangerous or more.”
“You can't really improve what you if you don't know what happened and that's where observability comes in. And then when you do improve, these LLMs are great, but like they're they're not robust at all.”
“We're putting the means of opportunity in the hands of people who wouldn't otherwise have been able to build this thing.”
“AI unlocks scale and context unlocks that trust and confidence in what you're seeing. So this is a long-winded answer, Ankur. You asked is this just shift left? I prefer to think of it as shifting down.”
“And we finally live in a world where that is true, which means like almost anyone has the ability to build.”
“Fundamentally, we're all becoming managers of agents. So, a lot of the code is being written by by AI.”
“So, we really want to make sure that these models are accessible for the ecosystem and there's this thing which we call the Gemmaverse.”
Chapters
73
Articles
11
Google Cloud is revolutionizing cross-platform development by bringing Dart, the language powering Flutter, to Firebase Functions. This pivotal update allows developers to write their entire application, from mobile and web frontend to serverless backend, in a single, high-performance language.
Google Cloud is fundamentally changing how developers build and manage AI agents with the introduction of Agent CLI and significant enhancements to the Agent Development Kit (ADK 2.0), pushing AI from experimental demos to robust, production-ready systems.
At Google Cloud Next, industry leaders Dave Elliott and Adi Osmany introduced the Gemini Enterprise Agent Platform, a comprehensive solution designed to tackle the pervasive challenge of moving AI agent prototypes into reliable, production-grade systems.
In an insightful session, Google Developer Expert Tomek Wierzchowski demystifies the process of building AI applications, sharing his journey from a multi-voice audiobook concept to a deployable solution. He emphasizes practical tools and a strategic mindset for navigating the fast-evolving AI landscape.
Google Cloud has introduced the Gemini Enterprise Agent Platform, a comprehensive solution designed to empower both developers and business users to build, secure, and share AI agents with unprecedented ease and control.
In a revealing session, Harrison Chase, CEO and co-founder of LangChain, highlighted a paradigm shift in AI agent development: the 'agent harness' layer is where the most significant performance gains are being made, often surpassing the impact of model weights.
At Google Cloud Next, Logan Kilpatrick, a key figure in AI Studio's development, shared insights into the platform's rapid evolution, emphasizing its role in democratizing software creation and pushing the boundaries of what's possible with AI.
At Google Cloud Next, Wiz and Google Cloud introduced a transformative approach to AI security, moving beyond traditional "shift left" paradigms to empower developers with integrated, AI-driven protection.
The Google Cloud Next '26 Developer Keynote delivered a wave of innovations, signaling a pivotal shift towards more accessible, powerful, and collaborative AI development. Leaders Sarah Kennedy and Ricky Ravenet highlighted key advancements designed to empower developers across all skill levels.
Michele Catasta, President and Head of AI at Replit, shared a compelling vision for the future of software development, asserting that the role of a developer is undergoing a fundamental transformation driven by AI.
Google DeepMind's recent launch of Gemma 4 has sent ripples through the AI community, demonstrating unprecedented success with its open-access models designed for deployment across a vast spectrum of devices.
Sessions
8
