ODSC AI Unboxed: Hands-On Lightning Sessions

Schedule

The time frames are only estimates and may vary according to how the class is progressing.

Sheamus McGovern: Event Introduction (10 minutes)

• Host Sheamus McGovern welcomes you to ODSC AI Unboxed.

Segment 1 (30 mins) Open-Source Models Are the Future of Enterprise AI with Ivan Lee, CEO at Datasaur:

As enterprises explore the potential of generative AI, they face a critical tension: leveraging cutting-edge models while ensuring security, compliance, and control. Open-source models provide transparency, flexibility, and a rapidly evolving ecosystem, but to meet enterprise needs, they must be deployed in secure environments where sensitive data never leaves company servers. Equally important, enterprises must adapt these models to their own proprietary data and workflows—training them on internal processes, legal frameworks, and domain-specific knowledge. This talk outlines why open-source models are uniquely positioned to meet these demands, and how organizations can build private, secure, and fine-tuned AI systems that drive real business value.

Discussion and Q&A with Sheamus McGovern and Ivan Lee

Ivan Lee graduated with a Computer Science B.S. from Stanford University, then dropped out of his master’s degree to found his first mobile gaming company Loki Studios. After raising institutional funding and building a profitable game, Loki was acquired by Yahoo. Lee spent the next 10 years building AI products at Yahoo and Apple and discovered there was a gap in serving the rapid evolution of Natural Language Processing (NLP) technologies. He built Datasaur to focus on democratizing access to NLP and LLMs. Datasaur has raised $8m in venture funding from top-tier investors such as Initialized Capital, Greg Brockman (President, OpenAI) and Calvin French-Owen (CTO, Segment) and serves companies such as Google, Netflix, Qualtrics, Spotify, the FBI and more.

Segment 2 (30 mins) Build Production-Ready Agents with Simple-to-Customize UIs with Sandi Besen, AI Research Engineer and Ecosystem Lead at IBM Research:

As organizations explore agentic AI, the challenge is no longer whether agents can be built, but how to move from proof-of-concepts to production-ready systems. In this session, we’ll walk through the design principles and practical considerations for developing robust AI agents at scale. We’ll examine key capabilities that matter in real-world deployments—such as observability, memory management, and tool access—and how integrated user interfaces can accelerate adoption across enterprise teams.

Discussion and Q&A with Sheamus McGovern and Sandi Besen

Sandi Besen is an AI Research Engineer and Ecosystem Lead at IBM Research, where she drives initiatives in research-to-product translation and ecosystem development. Her work focuses on turning cutting-edge research into scalable solutions, often in collaboration with the open-source community to accelerate innovation. Previously, as an Applied AI Researcher at Neudesic, she designed multi-agent AI systems and authored widely cited research papers and meta-analyses. She is also an active AI advisor to startups, guiding them from concept to deployment and aligning technical capabilities with strategic goals.

Break (5 mins)

Segment 3 (30 mins) Automating Massive Refactors with Parallel Agents with Robert Brennan, CEO of All Hands AI:

Today's agents are best at small, atomic coding tasks. Much larger tasks--like major refactors and breaking dependency updates--are highly automatable but hard to one-shot. A human needs to thoughtfully orchestrate agents, check their work, and unblock them to fully accomplish the task.

In this session, we'll discuss patterns for orchestrating large-scale code changes with swarms of agents and a human in the loop. We'll use OpenHands as an example, and show how the open source Agent SDK can be used to drive a fleet of agents, and even pass messages between agents.We'll also work through a concrete example: migrating an entire codebase from one React state management library to another.

Discussion and Q&A with Sheamus McGovern and Robert Brennan

Robert Brennan has been writing software for 15 years, with a focus on natural language processing and developer tools. He is currently the CEO of All Hands AI, the company behind OpenHands (formerly OpenDevin), a fully autonomous AI developer. Previously he was VP of Product Development at Fairwinds, ran a startup called Datafire, and worked as a Senior Software Engineer at Google.

Segment 4 (30 mins) A Practical Guide to LLM Evaluation with Michelle Yi, Co-Founder at Generationship:

As organizations race to integrate Large Language Models (LLMs) into products and workflows, the challenge of robust evaluation has become a bottleneck. Standard academic benchmarks often fail to predict real-world performance, leaving teams unsure if their models are reliable, safe, or truly effective for their specific use case and leaving open the question of business value. This gap between benchmark scores and operational utility introduces significant risks, from brand-damaging hallucinations and biased outputs to poor user experiences that erode trust and even being unable to justify ROI. This condensed session provides an overview of why evaluation matters, the types of evaluation needed for generative models, and a quick look at where evaluation is headed.

Discussion and Q&A with Sheamus McGovern and Michelle Yi

Michelle Yi is a technology leader and investor that specializes in AI and cloud computing. She has 20 years of experience in the technology industry, contributed to the original IBM Watson showcased on Jeopardy, and enjoys building and leading teams that develop and deploy AI solutions to solve real-world problems. Michelle is passionate about diversity, STEM education/careers for our minority communities, and serves on the board of Women in Data.

Break (5 mins)

Segment 5 (30 mins) Context Engineering: A State-of-the-Art Overview with Greg Loughnane and Chris Alexiuk, Co-Founders of AI Makerspace

In this talk, we break down how Context Engineering has achieved the next layer of abstraction over Prompt Engineering, and how this new layer is similar to the way that LangGraph abstracted LangChain. What is the new mental model that we should use for Context Engineering? What is the taxonomy that we should think about when reaching into our production AI Engineering toolkit? To accomplish this, we’ll consider all of the things that we can put in context, including instructions, developer prompts, user prompts, state, short and long-term memory, retrieved information, tools, definitions of required structured output, and more!

Discussion and Q&A with Sheamus McGovern, Greg Loughnane and Chris Alexiuk

Dr. Greg Loughnane is the Co-Founder & CEO of AI Makerspace, where he is an instructor for their AI Engineering Bootcamp. Since 2021, he has built and led industry-leading Machine Learning education programs. Previously, he worked as an AI product manager, a university professor teaching AI, an AI consultant and startup advisor, and an ML researcher. He loves trail running and is based in Dayton, Ohio.

Chris “The Wiz” Alexiuk is the Co-Founder & CTO at AI Makerspace, where he is an instructor for their AI Engineering Bootcamp. During the day, he is also a Developer Advocate at NVIDIA. Previously, he was a Founding Machine Learning Engineer, Data Scientist, and ML curriculum developer and instructor. He’s a YouTube content creator whose motto is “Build, build, build!” He loves Dungeons & Dragons and is based in Toronto, Canada.

Sheamus McGovern’s closing remarks (10 minutes)