From Base Model to Agent: Inside NVIDIA’s PostTraining Stack
with Aishwarya Srinivasan, Arvind Narayanamurthy
Modern AI capability is increasingly shaped after pre-training. This session breaks down the systems behind NVIDIA Nemotron, from SFT and RLVR to verifiers, reasoning control and infrastructure, so builders can understand how today’s reasoning and agentic models are actually trained, evaluated and improved.
Loading session access
What you will learn
1
Understand the Modern Post-Training Stack
See how SFT, RLVR and RLHF shape a base model into a capable assistant, reasoner and agent.
2
Design Rewards Models Can’t Game
Learn why strong verifiers are critical for RL, evals and reliable model improvement.
3
Choose the Right Model and Reasoning Budget
Learn when to use smaller vs. larger models and how reasoning depth impacts cost, latency and quality.
4
Think Like a Post-Training Systems Engineer
Understand how environments, rollout infrastructure and resiliency shape frontier model performance.
About this session
Modern AI capability is increasingly shaped after pre-training. This session breaks down the systems behind NVIDIA Nemotron, from SFT and RLVR to verifiers, reasoning control and infrastructure, so builders can understand how today’s reasoning and agentic models are actually trained, evaluated and improved.
Guest speaker
Chris Alexiuk
Product Research Engineer and Developer Advocate @ NVIDIA · NVIDIA
Chris Alexiuk is a Product Research Engineer and Developer Advocate at NVIDIA , working at the intersection of machine learning research, generative AI, and developer education. His work spans NVIDIA’s AI ecosystem, including Nemotron, post-training, open models, and production AI systems . He also writes for the NVIDIA Technical Blog, speaks at AI conferences, and co-authored The AI Engineering Bootcamp from Wiley.
The Gen Academy
Hosted by Aishwarya Srinivasan and Arvind Narayanamurthy.
Aishwarya Srinivasan
AI Advisor / AI Educator
Aishwarya Srinivasan is the Co-Founder of The Gen Academy and a Developer Relations Lead at Nebius. Her experience spans Fireworks AI, Microsoft for Startups, Google Cloud and IBM, where she worked with developers and businesses on practical AI and machine-learning solutions. She holds a postgraduate degree in Data Science from Columbia University and teaches AI engineering, applied machine learning, generative AI and LLMOps.
Arvind Narayanamurthy
Founder @ The Gen Academy | AI Solutions Architect | Ex - Adobe, Microsoft, IBM
Arvind brings a cross-company perspective on AI at scale, built across Microsoft, IBM, and Adobe, where he led data science teams and developed production-grade machine-learning solutions. He has built and deployed enterprise AI systems and focuses on translating AI concepts into practical, production-ready solutions. He holds a Master's from Carnegie Mellon University.
