Session Slides & Recordings

AIOS AI SuperUser Bootcamp

Mark Hinkle

Become a ChatGPT SuperUser & Build AI Agents for 2025

GenAI RAG Workshop

John Willis

The “GenAI RAG Workshop” is an immersive, full-day training designed for engineers seeking to harness the power of Retrieval-Augmented Generation (RAG) for AI-driven solutions. Participants will explore foundational concepts, including embeddings, vector databases, and advanced RAG techniques while delving into practical applications of tools like MongoDB Atlas Vector Search. Through a blend of lectures and hands-on labs, this workshop equips attendees with the skills to effectively design, implement, and evaluate RAG pipelines, focusing on optimizing AI systems for precision, scalability, and relevance. By the end of the session, participants will understand how to integrate RAG with cutting-edge vector database tools and frameworks to enhance AI-driven workflows.

Build GCP certification tutor with Gemini, LangChain and Vector Search

Rishab Kumar

Learn how to architect and implement an AI-powered study companion using Google Cloud’s Vertex AI with Gemini and Vector Search. We’ll dive into the technical implementation of embedding generation, vector storage, and prompt engineering to create a dynamic certification preparation system. This session will cover practical code examples, architecture decisions, and deployment patterns for building production-ready AI applications on Google Cloud.

Hallucinations, Prompt Manipulations, and Mitigating Risk: Putting Guardrails around your LLM-Powered Applications

Don Shin

Large Language Models (LLMs) offer a future where we can interact with computers naturally like we interact with humans. However amazing this potential, enterprise adoption is still hindered by the risk of hallucinations – the propensity of LLMs to respond inaccurately but in a voice of complete confidence. Also, LLMs by themselves can be easily manipulated by end-user prompts to respond in ways that can damage trust or create liability. In order for LLM-powered applications to be fully utilized in real-world production settings, risk must be mitigated through effective guardrail strategies.

You Don’t Need an AI Strategy, But You Do Need to Be Strategic About AI

Jessica Hall

There’s so much noise about creating an “AI strategy,” it’s easy to feel like you’re already behind. But here’s the thing: you don’t need an AI strategy or a data strategy. Those things need to serve your business strategy and that requires strategic thinking.

Cooking with Granite : Granite AI Models Workshop

BJ Hargrave

The Granite family of foundation models span an increasing variety of modalities, including language, code, time series, and science (e.g., materials) – with much more to come. This walk-through of the Granite Community’s Cookbooks for the IBM Granite models will give you hands-on experience using the Granite models (including those specialized for code and time-series) in short recipes. The session will start with simple use cases that demonstrate zero-shot inference with the code and time-series models. The recipes will grow more complex to cover data preparation, RAG, tool use, and agentic workflows. By the end of this workshop, you will be able to:

Testing GenAI Applications

Adrian Cole

Software engineers are increasingly using Generative AI to build applications like chatbots, content generation tools and even agents. Testing can be tricky, especially if against a billable account. It gets even trickier when you want your application to perform reliably.

Leveraging Pre-Trained Transformer Models for Protein Function Prediction

Tia Pope

Transformer-based models, such as ProtGPT2 and ESM, are revolutionizing protein sequence analysis by enabling detailed embeddings and advanced function prediction. This talk provides a hands-on introduction to using pre-trained open-source transformer models for generating protein embeddings and leveraging them for classification tasks. Attendees will learn to tokenize sequences, extract embeddings, and implement machine-learning pipelines for protein function annotation based on Gene Ontology (GO) or Enzyme Commission (EC) numbers. This session will showcase how pre-trained transformers can democratize access to advanced protein analysis techniques while addressing scalability and explainability challenges. After the talk, the speaker will provide a notebook to test basic functionality, enabling participants to explore the concepts discussed.

AI-Assisted Grant/RFP Writing Workflow: Scale Funding Efforts Without Scaling Costs

Will Mendoza

Our AI deployment assists our clients in generating the first draft of their grant proposals, taking into consideration donors’ formats and requirements. It uses the language and tone of our clients’ previous successes to create a draft that can then be evaluated and refined by the grant writer. This deployment has reduced costs and increased the number of grant proposals written by our clients.

Codename Goose: An Open Source Local AI Agent Built on MCP

Angie Jones

AI coding assistants are evolving fast, but most are limited to generating code. Codename Goose, an open source AI agent, takes it further by actually running commands, fixing code, and interacting with your development environment like a true assistant.

Practical Generative AI Observability: Metrics and Tools for Real-Time Monitoring

Tony Shan

As generative AI systems power ever more critical applications, ensuring the reliability, fairness, and performance of these systems demands robust observability frameworks. This presentation focuses on the emerging discipline of Generative AI Observability through a deep dive into strategies, methods, and best practices for real-time monitoring of generative systems. Attendees will learn metrics techniques to track key performance indicators such as output coherence, accuracy, and latency, while also gaining insights into how to detect and mitigate issues like bias, hallucination, and model drift. We’ll explore state-of-the-art observability tools designed for generative AI including those tailored for large language models, RAG frameworks, and multimodal systems. The discussion will cover innovation in monitoring the components in the pip

Responsible AI Considerations for Language Models

Sophia Rowland

AI is being used throughout our world in the systems and tools we use every day. This also means that AI is being used irresponsibly throughout our world. Irresponsible usage of AI incurs a cost, but it is not always a financial cost. As Language Models and Generative AI adoption increases, organizations open themselves to a new set of risks. In this session, Sophia Rowland will introduce potential pitfalls of Language Model adoption and how to address those pitfalls to build AI systems more responsibly.

Day 1 Networking Social

Monday, March 17 We’ll host a social immediately following onsite programming. Attendees can simply stick around and enjoy themselves after a long and engaging day of programming.

Check-In and Onsite Registration Opens

ATO.AI attendees can register onsite (for those not already registered) or check-in and grab credentials before heading into sessions.

Artificial Intelligence Needs Community Intelligence

Sriram Raghavan

Abstract coming soon!

Welcome Keynote

Welcome and Thank You’s/Acknowledgements

The Death of the Browser

Rachel-Lee Nabors

In ten years, Internet Browsers may be a nostalgic memory. As enterprises face mounting API costs and integration headaches, a new paradigm is emerging. The internet’s evolution from an open highway into a maze of walled gardens and monetized APIs has created significant challenges for businesses—but it has also set the stage for accessing and organizing the world’s information.

Don’t just talk to AI, do more with AI: how to improve productivity with AI agents

Sheng Liang

Abstract coming soon!

Open-Source GenAI vs. Enterprise GenAI: Navigating the Future of AI Innovation

Dr. Ruth Akintunde

This talk explores the critical differences between Open-Source Generative AI and Enterprise Generative AI, highlighting their respective strengths and challenges. Open-Source GenAI fosters innovation through community collaboration, accessibility, and adaptability, while Enterprise GenAI prioritizes security, scalability, and reliability. Key aspects such as cost, ethical considerations, and long-term sustainability are examined to understand their impact on AI development and deployment. Ultimately, the talk advocates for a hybrid approach, leveraging the best of both worlds to drive AI innovation forward.

DON’T PANIC: AI IS COMING – The Hitchhiker’s Guide to AI

Mark Hinkle

AI is coming of age, and much like discovering intergalactic travel, it’s equal parts thrilling and terrifying. Fears of job loss, doomsday scenarios, and bureaucratic AI overlords dominate the conversation—but I think the reality is far less apocalyptic and far more exciting. With the right guide, you can navigate this new universe, adapt, and even thrive. That’s what AllThingsOpen.AI is all about—building a community where people and businesses don’t just survive AI’s rise but flourish in it. So grab your towel, keep an open mind, and let’s explore the future—without the panic. Listen to Conference Co-Producer and publisher of the Artificially Intelligent Enterprise, Mark Hinkle, provide a vision on how AI will play out in our lives.

Prioritizing and Evaluating GenAI Use Cases

Ben Heller

Companies are inundated with GenAI Use Case opportunities and requests. Often the challenge is determining where to start: planning and understanding what the value, the actionability and the feasibility of each use case is key. This talk will walk through frameworks, questions, and provoking ideas on how AI engineers and AI users can implement at their own companies or potential companies they are aiming to find a market fit for. All of these components have a spirit of Responsible AI as there are no use cases without factoring this into the discussion.

Let’s Talk About LLM Evaluations

John Willis

Abstract coming soon!

​​Fine-Tuning Large Language Models with Declarative ML Orchestration

Shivay Lamba

Large Language Models used in tools like ChatGPT are everywhere; however, only a few organisations with massive computing resources are capable of training such large models. While eager to fine-tune these models for specific applications, the broader ML community often grapples with significant infrastructure challenges.

Let’s Create a GitHub Copilot Extension!

Nick Taylor

Get hands-on in this talk where we’ll create a GitHub Copilot Extension from scratch.

A truly community-based Open Source LLM and what it can do for your business and the open community as a whole

JJ Asghar

What if I told you how we’re distributing AI models is fundamentally broken? Most people out there leverage the ChatGPTs of the world and just ask it for some simple tasks leveraging Prompt Engineering. That’s all well and good, but when you want to bring AI into your business processes, do you want to give your secret sauce to another company to train their models? Probably not.

The Future is Non-Deterministic

Michael Donovan

Programming has been a deterministic art form since the dawn of computer science. This has influenced the way we solve problems, the type of user experiences we build, the architectures we create, and the primitives we use. Hidden underneath the hype of GenAI chat interfaces and image generation tools, lies the true next frontier of computing that GenAI is giving us. Agentic Workflows, made possible through the “tool calling” interface available in LLMs, creates an entirely new way to solve problems through non-deterministic solutions. In this talk I will explore how non-deterministic solutions, powered by Agentic Workflows, can change the way we build everything from developer tools to automated tests to microservices.

How I created Bookshelf : An Open Source AI-Powered Personal Knowledge Base

Ash Tewari

Session: How I created Bookshelf : An Open Source AI-Powered Personal Knowledge Base Join me as I walk through the design and engineering process behind BookShelf, a generative AI-powered application that serves as a personal knowledge base built from private content. This session focuses on the practical aspects of creating a Retrieval-Augmented Generation (RAG) application, offering actionable insights for developers and architects.

Beyond Single-Source RAG: Cross-Stack AI Assistants

Kevin Blanco

In 2023, 68% of enterprises reported their AI initiatives were hampered by data accessibility issues across different platforms.

AI-assisted Open Source Issue Management

Arundeep Nagaraj

As open source projects scale, managing the growing volume of issues while maintaining high-quality customer interactions becomes increasingly challenging. This talk presents a novel approach that leverages AI to assist in issue management in open source repositories. We demonstrate how Gen AI can be used to help in triaging issues as well as analyzing sentiment in real-time, and provide signals that streamline the triage process to reduce frustration from all stakeholders. By aggregating and analyzing issue patterns and user feedback, we can further provide actionable insights to product and engineering teams, enabling data-driven prioritization and feature development.

Safeguarding Innovation: Navigating Patent Challenges in the Expanding AI Landscape

Keith Bergelt

It is an understatement to say that the Artificial Intelligence (AI) market is large, growing, and increasingly the focus of businesses, technology providers and investors. Leveraging large language model (LLM) machine learning has brought incredible investment, growth, innovation and adoption of AI. The global AI market, valued at over $136 billion in 2022, is projected to grow at a compound annual growth rate (CAGR) of 37.3%, between 2023 to 2030 – when it is estimated to reach approximately $1.7 trillion annually.

AI Observability

Wade Moore

Modern observability platforms are rapidly integrating the latest advancements in AI to help engineers resolve incidents, investigate issues, and develop faster. Observability is inherently a data problem, and with vast amounts of data already flowing through these platforms, the opportunity to leverage AI is unprecedented.

Understanding Vector Databases

Jessica Garson

Vector databases have established themselves as the new semantic search norm, prioritizing the context and meaning of content over the conventional keyword-based approach. This talk will explore the traditional “Classic” search, highlight its constraints, and provide an overview of vector databases. This talk is aimed for developers who may have some experience with machine learning but looking to skill up.

Coding with AI, Revolution or Evolution

Brandon Mathis

People say coding will never be the same after AI. But is this the first time something like this has happened in our industry? What did coders say when the first compiler was built? Is copying code from ChatGPT the same as copying code off Stack Overflow? Will AI take my job? In this talk I will challenge preconceived notions of how we as engineers should use AI and how will it truly impact our profession.

Rolling out Enterprise AI: Tools, Insights, and Team Empowerment

Calvin Hendryx-Parker

C-level executives, directors, and product managers face the complex task of integrating AI into existing systems while addressing challenges related to data quality, scalability, interoperability, ethics, skill gaps, and security compliance.

AI Governance: Mitigating Risks in the Enterprise

Melissa Nysewander

With AI is gaining traction in every industry, with new uses popping up every day, companies need to also adopt new AI policies to ensure they are mitigating the potential risks. These risks range from meeting regulatory and compliance needs, to ensuring the ethical use, to ensuring that companies are getting the proper return on their investment. Here, I will discuss the various factors required for a comprehensive AI governance program.

Accelerating GenAI Business Value

Here’s a draft abstract focused on Generative AI deployment at Labcorp:

Leveraging Knowledge Graphs for RAG: A Smarter Approach to Contextual AI Applications

David vonThenen

In the ever-evolving field of AI, retrieval-augmented generation (RAG) systems have become critical for delivering high-quality, contextually relevant answers in applications powered by large language models (LLMs). While vector databases have traditionally dominated RAG applications, graph databases, specifically knowledge graphs, offer a transformative approach to contextual AI that’s often overlooked. This approach provides unique advantages for applications requiring deep insights, intelligent search, and reasoning over both structured and unstructured sources, making it ideal for complex business scenarios.

Search in the AI era: how semantic search and agentic AI are revolutionizing information discovery

Jon Reifschneider

One of the most significant impacts of large language models (LLMs) is the way they are changing how we discover information. Traditional search methods are evolving as AI introduces capabilities like natural language understanding, semantic search and agent-based search, allowing for more intuitive and personalized search experiences. We’ll explore new technologies and case studies that showcase how AI-driven search systems can decipher user intent, deliver context-aware results, and continually learn from interactions to enhance accuracy over time. Join us to uncover the transformative impact of AI on search technology and discuss the opportunities and risks that lie ahead in AI-powered information retrieval.

Sneaky Ways to Integrate GenAI for Real-World Impact

Ben Ilegbodu

AI is no longer just a buzzword; it’s a transformative force poised to reshape how we interact with technology. Everyone is trying to create the latest game-changing AI-driven app, but how can we developers leverage GenAI without becoming full-fledged AI experts or overhauling our existing applications?

Building AI Agents and Assistants for DevOps Workflows

Lee Faus

In this session we will show the differences between agents and assistants with real world examples. We will demonstrate integrating assistants into open source projects like continue.dev using open source LLMs served up from Ollama. We will demonstrate how assistants can remove toil from different personas in the software supply chain from product managers to release managers. We will finally demonstrate how to stitch assistants and agents together using LangChain, Bee, and LlamaIndex using RAGs to streamline communication and elevate collaboration using these technologies.

Gen AI: AI Agents – Making LLMs work together in an organized way

Brent Laster

AI Agents are combinations of LLMs, tools, and custom roles that can autonomously perform tasks and make decisions based on context and user input. Multiple agents can be managed together to cooperatively handle individual tasks that are part of a larger project to accomplish an overall goal.

Unleashing the Power of LLMs: Revolutionizing Drug Safety and Business Automation with Multi-Agent AI

Venkateswara Rao Chalamalasetti

In this presentation, we explore how Large Language Models (LLMs) are transforming drug safety and business automation. We’ll showcase case studies where AI and LLMs have advanced drug safety science, including enhanced user interactions through chatbots and converting natural language queries into database searches. Operating within a regulated environment, we’ve developed guidelines to prevent hallucinations in translation tasks, ensuring trustworthy AI implementation.

Hacking AI – How to Survive the AI Uprising

Gant Laborde

While AI can recognize faces under masks, track purchases, and watch you sleep, there’s still hope when the Great Robot Revolution kicks off! Hacking software was cool in the 90s, but hacking data against AI will be cool tomorrow. We’ll discuss some of the mechanics for data bias, image perturbation, and jailbreaking that are emerging as their own mechanics against machines. No prior AI knowledge is needed for this talk, everyone will follow along and learn!

Unleashing the Power of Vector Databases for AI-Driven Applications

Venkat Venkatakrishnan

Vector databases are emerging as the cornerstone of next-generation data management, enabling seamless storage, indexing, and retrieval of dense vector embeddings derived from AI models. In an era dominated by artificial intelligence and machine learning, the ability to handle high-dimensional data efficiently is critical and a game-changer.

Scaling Large Models with Model & Data Parallelism: Techniques, Tradeoffs, and Best Practices

Shashank Kapadia

Discover how to train and serve massive AI models efficiently by leveraging both model and data parallelism. In this session, we’ll explore how to partition large models across GPUs and distribute data for optimal throughput, diving deep into practical setup details and performance benchmarks. We’ll also address the key tradeoffs—such as latency vs. resource usage—and show how to tailor parallelization strategies to different AI tasks, going beyond transformers into computer vision and more. By the end, you’ll have a holistic understanding of how to design and deploy parallelized workflows that balance accuracy, speed, and infrastructure costs, enabling you to scale AI solutions effectively in real-world scenarios.

Scaling Success: Transforming PoC into Large-Scale Impact through Iteration

Dhivya Nagasubramanian

Scaling a Generative AI (GenAI) project from a Proof of Concept (PoC) to a large-scale deployment is a journey that demands careful planning and an iterative approach. In this talk, I’ll Walk you through how to take a PoC and scale it incrementally, learning from each stage to ensure that the final solution meets real-world demands. I’ll highlight the importance of refining your model based on early feedback, optimizing performance at every step, and expanding capabilities as you go. By testing in controlled environments, gathering real-time data, and using agile methodologies, we can mitigate risks and ensure smoother transitions as the project grows. This iterative approach not only makes scaling manageable but also helps create a more robust, adaptable GenAI solution that can deliver meaningful impact at scale.

A Developer’s Adventures in AI Wonderland

Burr Sutter

This session is targeted at enterprise application developers and hands-on architects who are interested in building LLM-infused enterprise applications. While we will have some fun, this will be a practical, demo-oriented, learn by example, and code-oriented session.

Security and AI: The Strategic Imperative

Katherine Druckman

The rapid advancement of artificial intelligence (AI) presents unprecedented opportunities, but also introduces critical security and privacy challenges. Organizations face increasing pressure to safeguard sensitive data in AI-driven applications, as evidenced by a Cisco survey revealing that 91% of security and privacy professionals recognize the need for greater transparency and customer reassurance regarding AI’s data usage. Furthermore, Gartner reports that 40% of organizations have already experienced an AI-related privacy breach, highlighting the urgency of this issue.

Closing Keynotes

Keynote speakers and talks to be announced soon!