Subvolts take Coding and reinforcement learning may unlock human-level AI faster than other methods. Sep 24, 2025 - Developers
- Learn how AI advancements, particularly in coding and agentic systems, will change software creation and your role.
- Business
- Understand how AI agents and improved coding tools could accelerate product development and innovation cycles.
This interview provides a detailed look at a specific technical approach to achieving AGI from a company focused on that goal.
#289 Eiso Kant: How Reinforcement Learning and Coding Could Unlock Human-Level AI
Audio streams from Craig S. Smith. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Eye on AI episode #289 features Eiso Kant, co-founder of Poolside, discussing how reinforcement learning and software development could accelerate the path to human-level AI. Kant explains Poolside's approach using reinforcement learning from code execution (RLCF) and argues that coding provides an ideal training ground for intelligence. The conversation explores agentic AI systems and their potential to transform software creation.
This episode explores how teaching AI to write and test code, like a developer, could be the quickest way to create truly intelligent machines. It's like giving AI a job where it learns by doing and fixing its mistakes, moving it closer to human-like understanding.
This episode is for anyone interested in the future of artificial general intelligence (AGI) and how it might be achieved. Kant, from Poolside, offers a specific technical perspective on using reinforcement learning with code execution as a pathway to AGI. This connects to current frontier lab efforts in developing more capable and autonomous AI systems, particularly in areas like coding assistance and agentic behavior, which are active research areas for Google Gemini, OpenAI, and Anthropic.
- HostCraig S. Smith
- GuestEiso KantCo-Founder, Poolside
- Episode Title: #289 Eiso Kant: How Reinforcement Learning and Coding Could Unlock Human-Level AI
- Show: Eye on AI
- Guest: Eiso Kant, Co-Founder of Poolside
- Length: 54 minutes
- Method Discussed: Reinforcement Learning from Code Execution (RLCF)
How do we get from today's AI copilots to true human-level intelligence? In this episode of Eye on AI, Craig Smith sits down with Eiso Kant, Co-Founder of Poolside, to explore why reinforcement learning + software development might be the fastest path to human-level AI. Eiso shares Poolside's mission to build AI that doesn't just autocomplete code — but learns like a real developer. You'll hear how Poolside uses reinforcement learning from code execution (RLCF), why software development is the perfect training ground for intelligence, and how agentic AI systems are about to transform the way we build and ship software. If you want to understand the future of AI, software engineering, and AGI, this conversation is packed with insights you won't want to miss. Stay Updated: Craig Smith on X:https://x.com/craigss Eye on A.I. on X: https://x.com/EyeOn_AI (00:00) The Missing Ingredient for Human-Level AI(01:02) Eiso Kant's Journey(05:30) Using Software Development to Reach AGI(07:48) Why Coding Is the Perfect Training Ground for Intelligence(10:11) Reinforcement Learning from Code Execution (RLCF) Explained(13:14) How Poolside Builds and Trains Its Foundation Models(17:35) The Rise of Agentic AI(21:08) Making Software Creation Accessible to Everyone(26:03) Overcoming Model Limitations(32:08) Training Models to Think(37:24) Building the Future of AI Agents(42:11) Poolside's Full-Stack Approach to AI Deployment(46:28) Enterprise Partnerships, Security & Customization Behind the Firewall(50:48) Giving Enterprises Transparency to Drive Adoption
- Reinforcement Learning from Code Execution (RLCF)
- Eiso Kant
- AI learning by running code and getting feedback on results.
- Agentic AI
- Eiso Kant
- AI systems that can act independently to achieve goals.
55 min42 min51 min34 minUkraine's Secret Weapon: The Points System Winning the Drone War | Andrii Hrytseniuk
WithAndrii Hrytseniuk
Same guests, other shows
Questions people ask
- What is Reinforcement Learning from Code Execution (RLCF)?
- RLCF is a method where AI learns by executing code and receiving feedback on its performance, similar to how developers learn from running their programs.
- Why is software development a good training ground for AI intelligence?
- Software development offers a structured environment with clear objectives, feedback loops, and complex problem solving, which are ideal for training advanced AI.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Craig S. Smith.