Subvolts take Learn how Google built Gemini's coding power and what's next for programming. Jun 16, 2025 - Developers
- Learn strategies for AI-assisted coding, handling large projects, and the future of programming languages.
- Business
- Understand how AI is transforming software development and the tools that will shape future products.
A solid interview with Google's leads on Gemini's coding features, offering detailed insights into AI development.
Building Gemini's Coding Capabilities
Audio streams from Google AI. Your place is remembered on this device.
From Subvolts
Summary and analysis by Subvolts editorial, AI-assisted.
Google AI: Release Notes features Connie Fan and Danny Tarlow discussing Gemini's coding capabilities. They cover early goals, the impact of 'vibe coding,' strategies for large codebases, and the future of programming languages. The episode explores how Gemini became a leading AI coding model.
This episode explains how Google created its AI that helps write computer code. It covers new ideas like 'vibe coding' and how AI might change programming languages in the future, which could affect how software is made.
This episode offers insight into the development of advanced AI coding tools, a key area for all frontier labs. It's for developers and AI researchers interested in how models handle complex code and future programming paradigms. Google's work here directly competes with efforts from OpenAI, Anthropic, and Meta in the AI developer tools space.
- HostLogan Kilpatrick
- GuestConnie FanProduct Lead for Gemini's coding capabilities, Google
- GuestDanny TarlowResearch Lead for Gemini's coding capabilities, Google
- Episode Title: Building Gemini's Coding Capabilities
- Show: Google AI: Release Notes
- Length: 60 minutes
- Guests: Connie Fan, Danny Tarlow
- Topics Covered: Early coding goals, 'vibe coding', large codebases, future of programming languages
Connie Fan, Product Lead for Gemini's coding capabilities, and Danny Tarlow, Research Lead for Gemini's coding capabilities, join host Logan Kilpatrick for an in-depth discussion on how the team built one of the world's leading AI coding models. Learn more about the early goals that shaped Gemini's approach to code, the rise of 'vibe coding' and its impact on development, strategies for tackling large codebases with long context and agents, and the future of programming languages in the age of AI.
Watch on YouTube: https://www.youtube.com/watch?v=jwbG_m-X-gE
Chapters:
0:00 - Intro
1:10 - Defining Early Coding Goals
6:23 - Ingredients of a Great Coding Model
9:28 - Adapting to Developer Workflows
11:40 - The Rise of Vibe Coding
14:43 - Code as a Reasoning Tool
17:20 - Code as a Universal Solver
20:47 - Evaluating Coding Models
24:30 - Leveraging Internal Googler Feedback
26:52 - Winning Over AI Skeptics
28:04 - Performance Across Programming Languages
33:05 - The Future of Programming Languages
36:16 - Strategies for Large Codebases
41:06 - Hill Climbing New Benchmarks
42:46 - Short-Term Improvements
44:42 - Model Style and Taste
47:43 - 2.5 Pro’s Breakthrough
51:06 - Early AI Coding Experiences
56:19 - Specialist vs. Generalist Models
- vibe coding
- Danny Tarlow
- A less formal, more intuitive approach to writing code, possibly influenced by AI suggestions.
- long context
- Danny Tarlow
- The ability of an AI model to remember and process a very large amount of information at once, like an entire codebase.
- agents
- Danny Tarlow
- AI systems designed to perform specific tasks or actions, often in response to user input or environmental changes.
27 minKoray Kavukcuoglu on frontier models, coding agents, and building AGI
WithKoray Kavukcuoglu
39 minIntroducing Gemini Robotics 2
WithCarolina Parada, Stuart Bowers, Kanishka Rao, Jie Tan
42 minGemini co-leads on project origins and what's next
WithJeff Dean, Koray Kavukcuoglu, Noam Shazeer, Oriol Vinyals
40 min
40 minQuestions people ask
- What are the early goals for Gemini's coding capabilities?
- The episode discusses the early goals that shaped Gemini's approach to code development.
- How does Gemini handle large codebases?
- Strategies for tackling large codebases with long context and agents are covered in the discussion.
Page generated Oct 19, 2026. Summaries are Subvolts' own; the audio, artwork and notes belong to Google AI.