The AI engineering landscape is evolving rapidly, and staying ahead means not just learning new tools but mastering the art of building scalable, safe, and effective AI solutions. At ODSC East 2025, we’re offering a curated set of sessions designed specifically for AI and data engineers who want to deepen their technical expertise and lead the next generation of AI applications.

Whether you’re passionate about autonomous agents, building secure GenAI systems, or harnessing small, efficient models, these sessions will give you the edge. Here’s a preview of some can’t-miss talks and workshops.

AI Software Engineering Agents: What Works and What Doesn’t

Robert Brennan, CEO of All Hands AI

This session teaches AI and data engineers how to set realistic expectations for software development agents like OpenHands (formerly OpenDevin), helping them leverage these tools effectively to boost productivity while avoiding noise and technical debt.

Agentic AI in Action: Build Autonomous, Multi-Agent Systems Hands-On in Python

Dr. Jon Krohn, Host of the SuperDataScience Podcast |Edward Donner, Co-founder and CTO of Nebula.io

This hands-on workshop gives AI and data engineers a deep dive into designing, building, and deploying autonomous multi-agent systems in Python using frameworks like LangChain and CrewAI, equipping them with both conceptual foundations and real-world coding experience to lead in the fast-evolving agentic AI landscape.

Effective AI Decision Support: Overcoming Both Human and AI Fallibility

Finale Doshi Velez, PhD, Professor at Harvard University

This talk offers AI and data engineers research-backed principles for designing AI decision-support systems that improve human-AI collaboration, covering how context-sensitive explanations and user-tailored trust strategies can lead to better, safer outcomes across industries beyond healthcare.

Implementing High-Quality and Cost-Efficient AI Applications with Small Language Models

Julien Simon, Chief Evangelist at Arcee.ai

This session teaches AI and data engineers how to overcome privacy, compliance, and cost challenges by adapting small open-source language models (SLMs) using tools like Arcee AI SLM, DistillKit, Spectrum, and MergeKit, enabling them to build efficient, domain-specific models for advanced workflows like model routing and agentic systems.

Beyond the Prompt: Architecting Reliable Enterprise LLM Agents

Vivek Muppalla, Director of AI Engineering  at Cohere

This session guides AI and data engineers through the complex decision-making needed to build scalable, safe enterprise-grade LLM agents, showcasing real-world strategies like human-in-the-loop safety, robust tool design, evaluation criteria, and synthetic data training—centered around a customer support use case.

Guardrails in Generative AI Workflows via Orchestration

Evaline Ju, Senior Software Engineer at IBM | Gaurav Kumbhat, Software Architect at IBM

This session empowers AI and data engineers to integrate scalable, production-ready guardrails into generative AI workflows using an open-source orchestrator component, helping them detect and mitigate biases, privacy leaks, and inappropriate content across multiple model runtimes and modalities.

Gen AI in Asset Management: Use Cases and Lessons from Building an Industry-Specific Copilot

Yu Yu, PhD, Director of Data Science at BlackRock

This session offers AI and data engineers valuable insights into how asset management firms evaluate and build GenAI solutions, including real-world strategies for scalable co-pilot development and key factors influencing the buy-versus-build decision process.

Build with AG2: Open-Source AgentOS

Chi Wang, PhD, Founder of AutoGen (Now AG2) and Senior Staff Research Scientist at Google DeepMind

This session empowers AI and data engineers to master agent-oriented programming by building production-ready, multi-agent systems that integrate models from OpenAI, Anthropic, Gemini, and others—unlocking deep research, customer service, and software development applications across industries.

Building Effective AI Agents with AG2

Qingyun Wu, PhD, Founder of AG2

This talk equips AI and data engineers with practical skills for building scalable, multi-agent systems using AG2 (formerly AutoGen), covering foundational design principles, collaboration patterns, and real-world implementation strategies to accelerate agent-based solution development.

Achieving Reliable and Responsible Enterprise AI Solutions via Coordinated Monitoring, Evaluation, and Governance

Dr Helen Gu, Founder and CEO of InsightFinder Inc.

This talk helps AI and data engineers tackle the reliability and governance challenges of enterprise-scale Composite AI systems by introducing intelligent monitoring, evaluation, and governance strategies, backed by real-world case studies to enhance trust, accountability, and system performance.

Predict & Prescribe: Combining Forecasts with Optimized Plans

Ryan O’Neil, PhD, CTO at Nextmv

This session teaches AI and data engineers how to combine machine learning with mathematical optimization to handle uncertainty in real-world business decisions, offering practical techniques for integrating predictive models into optimization workflows and creating more resilient, flexible solutions.

Transform Notebooks into Scalable Pipelines

Kevin Gunn, PhD, Lead Data Scientist at Plymouth Rock Assurance

This session shows AI and data engineers how to transform Jupyter notebook workflows into scalable, cost-efficient ML pipelines using Directed Acyclic Graphs (DAGs) in cloud environments like AWS, enabling faster development, better resource management, and streamlined production deployment.

Building the Backbone: Scalable LLMOps & MLOps to Enable AI Integration Across Teams

Pablo Vega-Behar, Director of Machine Learning Engineering at Fitch Group

This session provides AI and data engineers—and business leaders—with strategies for scaling AI operations enterprise-wide, showcasing how Fitch Group’s reusable LLMOps and MLOps components streamline AI integration, reduce deployment time, and enable non-specialist teams to incorporate AI without major headcount increases.

The A to Z of Building AI Agents

Apoorva Joshi, Senior AI Developer Advocate at MongoDB

This hands-on tutorial teaches AI and data engineers how to build fully functional AI agents from scratch, covering core concepts like reasoning, memory, and tool use, with a live coding walkthrough using LangGraph, MongoDB, Claude 3.5 Sonnet, and Hugging Face models.

Why You Can’t Miss These Sessions

The future of AI engineering belongs to those who can build systems, not just models. Whether it’s learning to coordinate multi-agent architectures, hardening AI pipelines, or mastering the real-world constraints of scaling enterprise AI, the sessions at ODSC East 2025 will equip you with skills that organizations urgently need today—and will prize even more tomorrow.

If you’re serious about taking your AI engineering career to the next level, these are the conversations you need to be part of. Sign up using this link for an additional 10% discount!