MLcon San Diego 2026

LLMs Under the Hood - Workshop In

Barry Stahl, AZNerds.net

this workshop, we’ll take a deep technical journey into Large Language Models (LLMs), focusing on three transformative aspects: Transformer architecture, embeddings, and Retrieval Augmented Generation (RAG).This seminar focuses on the building blo

Welcome and Opening Keynote of MLcon San Diego 2026

Sebastian Meyen, Software & Support Media GmbH

Welcome and Opening Keynote of MLcon San Diego 2026 by the Program Chair Sebastian Meyen.

Beyond Microservices: Migrating from Service-Oriented to Agent-Oriented Architecture Service-oriented architecture gave us scalability, but Agent-Oriented Architecture (AOA).

Andrew Miller, QuCelerate

This talk is for teams considering the shift, showing what really changes, what stays the same, and what breaks. You'll learn how AOA differs

Control, Comprehension, Context: DevOps in the Agentic Era Agentic AI doesn't just accelerate development — it breaks three things ops has always depended on: control over what.

Cory O'Daniel, Massdriver

Inside Model Reasoning: Fine-Tuning in Action What if I told you that advanced "thinking" AI models are just regular models that learned a new trick? In

Barry Stahl, AZNerds.net

this session, we'll use Azure Foundry to demonstrate the transformative power of fine-tuning by demonstrating a model fine-tuned using episodes

Teaching Machines to See and Read: The Science and Applications of Multimodal AI Multimodality represents one of AI's most fascinating and business-actionable frontiers, yet its.

Amit Bendor, Unity

this session, we'll explore multimodal architectures, ranging from CLIP to cutting-edge VLMs, and discover practic

Built-in AI & WebMCP: The AI Revolution Right in Your Browser More and more developers are looking to integrate Generative AI features into their applications. Until now, this.

Christian Liebel, Thinktecture AG

Stop Guessing, Start Measuring: Evaluations with Pydantic AI Your AI agent works well in demos, but how does it perform in real use? Without proper evaluation, you're taking a.

Andrew Miller, QuCelerate

This talk will show you how to use Pydantic AI's evaluation framework to measure key aspects: accuracy, consistency,

LLM-First UI: A Type-Driven Approach to AI-Generated Interfaces AI assistants promise to revolutionize how users interact with software, but there’s a hidden problem, how do you.

Navya Agarwal, Adobe

When an LLM generates arbitrary UI code, you get framework ve

The LLM Smörgåsbord Believe it or not, there are already over two million LLMs available on Hugging Face, and they don’t even include the public-facing GPTs like ChatGPT, Claude,.

John Davies, Incept5

Rage Against the Python Machine: Building a New Generation of Javascript ML When we hear 'AI' in 2026, we tend to think chatbots and API calls to massive Large Language Models..

Laurie Lay, Ippon Technologies

This talk expl

The Recipe for Relevance: Human‑Centered AI and Agentic Learning to Scale Skills AI frees time but will only deliver value when paired with human judgment, emotional.

Mary Tafuri, Sprinklr

This talk shows how agentic learning—assessment-driven, AI‑personalized pathways that assign learners ownership of their progress—re-ce

Zero Friction, Zero Cost: The Agentic Workflow for Azure Architects The industry is buzzing about "Vibe Coding" and

Saul Patino, CBTS

the rise of expensive, specialized AI editors that promise autonomous engineering. But do you really need a separate, paid subscription to unlock agentic capabilities?This session challenges the "pay-to

Building Action-Taking AI Agents: Reliability Lessons from Real Systems AI agents that generate or refactor code behave in predictable ways because their output remains text.

Adam Bell, Pulumi

When an agent begins taking real actions in external systems, the situation changes. Actions such as updating cloud infrastructure, running databas

Generating Playwright Tests with MCP and Playwright Agents for Intelligent QA Automation As software teams adopt AI copilots, conventional test automation struggles to keep pace..

Twinkle Joshi, IQGeo

this session, we’ll explore how Model Context Protocol (MCP)servers and Playwright Agents can work together to dynamically generate, validate, and self-heal

[Happy Hour – Incident Horror Stories in CloudOps: A Private, Peer Sharing Session with Cloud and Platform Leaders [ONSITE ONLY] Every team has at least one story they don't post.

Joe Karlsson, CloudQuery

Is DevOps doomed? With the arrival of AI-supported development tools everything in the software industry and the open source community is changing rapidly. Non-experts and.

Manfred Moser, Chainguard

AI as a Cog in the Machine: Transforming Unstructured Data into Competitive Advantage Most organizations sit on mountains of text, PDFs, emails, chats, images, and call.

Apurva Misra, Sentick

This session shows how to treat AI not as “the product,” but as a reliable cog insi

From LLMOps to AgentOps: Building and Evaluating Production-Ready AI Agents LLMOps enabled teams to productionize language models, but

Debjyoti Paul, Amazon

agentic systems introduce new challenges: multi-step reasoning, tool use, and autonomous decision-making. Traditional evaluation approaches—focused on single outputs—break down in these workf

Agents and MCP Servers: Are the Electric Sheep Safe? We have a new AI attack service. MCP servers are everywhere, and they are the new attack surface. Can the MCP server help.

Evaluating Agentic LLM Apps: Beyond Vibes Agentic AI has captured the industry's attention. As more people experiment with LLMs, they develop intuition aka vibes around whether.

Rushabh Mehta, Meta

Inside AI Agents: Debugging, Auditing, and Explaining Autonomous Decisions As AI agents become more autonomous, understanding their behavior becomes as important as their.

Debjyoti Paul, Amazon

How Developer Platforms Fail (And How Yours Won’t) Platform engineering promises so much, from developer productivity to better operational efficiency, and yet the Trough of.

Russell Miles, ClearBank

Breaking down an LLM for its parts AI has been around for many years. We’ve been using text-to-speech and speech-to-text for over a decade. Siri, Alexa, etc., cameras could.

John Davies, Incept5

Run LLMs, image, and speech models in the browser — no cloud required You don’t need a cloud subscription or backend infrastructure to bring powerful AI features into your web.

Maximiliano Firtman, firt.dev

AI-Driven Observability with MCP

Nitish Mane, Philo

The talk demonstrates how Platform and SRE teams can leverage AI to query metrics, explore logs, and reason about alerts using natural language. We will walk through practical MCP-based integrations, including Grafana MCP for metrics and dashboards,

Not Just Code - Characters: The Art of Human Oriented Programming Software has always been about logic. But what if your next system needed empathy? In

Nir Kaufman, Tikal

this talk, we’ll explore Human Oriented Programming (HOP) - a new way to design multi-agent systems by giving them human roles, motivations, and personalities We’ll

AI Literacy for Developers: Thinking Clearly While AI Writes the Code Synopsis: AI-assisted coding tools are changing what it means to be a developer.For decades, software.

Russell Miles, ClearBank

Full-Stack AI Engineering in TypeScript Bootcamp Day 1 Most AI tutorials stop at the "Hello World" of conversational bots. This intensive training takes you far beyond the.

Nir Kaufman, Tikal

Hands-On Bootcamp: Vibe Coding & GenAI Development Day 1 Master GenAI with ex-Visa Chief Architect John Davies. Hands-on coding in Python, Java, more. Build RAG, agents, and.

John Davies, Incept5

Strategic AI Workshop for Leaders An AI workshop for leaders is an immersive and practical program designed to equip executives and senior professionals with a comprehensive.

Garima Bajpai, Canada DevOps Community of Practice

The workshop begins by demystifyin

Two Day Mastering Platform Engineering – From Design to Value - Day 1 This two-day bootcamp with Russ Miles focuses on on the understanding of modern platform engineering and.

Russell Miles, ClearBank

You’ve built an Agent, now what? A Practical Guide to Evaluating, Debugging & Improving AI Agents In this hands-on workshop, participants will build a practical AI agent step by.

Apurva Misra, Sentick

Full-Stack AI Engineering in TypeScript Bootcamp Day 2 Most AI tutorials stop at the "Hello World" of conversational bots. This intensive training takes you far beyond the.

Nir Kaufman, Tikal

Hands-On Bootcamp: Vibe Coding & GenAI Development Day 2 Master GenAI with ex-Visa Chief Architect John Davies. Hands-on coding in Python, Java, more. Build RAG, agents, and.

John Davies, Incept5

Two Day Mastering Platform Engineering – From Design to Value - Day 2 This two-day bootcamp with Russ Miles focuses on on the understanding of modern platform engineering and.

Russell Miles, ClearBank