00:00:01 - Introducing John Cropezi
00:00:35 - Getting Started with Programming
00:06:42 - Developer Tools at Jane Street and the Case for Types
金尊
00:15:00 - Developer Tools at Jane Street and the Case for Types
agaduhan
00:06:42 - Developer Tools at Jane Street
00:11:32 - Lessons from Ruby and OCaml
00:14:51 - AI, Types, and Codebase Coordination
00:14:51 timestamp is transcript, sentence begins at 14:51 precisely. But max 3. Need topics key: origin, career/choosing Jane Street, Ruby/OCaml type discussion through AI. Better timestamps exact and <=3. Chapter 1 00:00:27 conversation background; 2 00:04:26 professional trajectory, 3 00:06:42 Jane Street developer tools then types later? Could choose 00:11:32 because substantive topic shift, but omit Jane Street discussion? Need max3. 00:00:27 origin, 00:06:42 developer tooling specialization, 00:11:32 Ruby vs OCaml and types. Exact formatting hh:mm:ss timestamps appearing transcript are bracket with 00:00:27 etc.00:00:27 - Getting Started with Programming
00:06:42 - Building Developer Tools at Jane Street
00:11:32 - Ruby, OCaml, and the Value of Types
00:15:02 - Language Choice and GitHub’s Performance
00:17:40 - Maintainability, Types, and Simplicity
00:22:14 - From Code Search to AI Tools and Infrastructure
00:30:02 - Evaluation and Model Harnesses00:15:02 - Language Choice and GitHub’s Performance
00:17:40 - Maintainability, Types, and Simplicity
00:22:14 - From Code Search to AI Tools and Infrastructure
00:30:05 - Challenges of Evaluating LLMs
00:34:00 - Security and Model Permissions
00:40:25 - Building and Refining Sandboxes
00:45:07 - Expanding Sandbox Permissions for AI Agents
00:48:13 - AI Users and Attribution
00:53:05 - From Tool Calling to Distributed Agentic Coding
01:00:08 - Distributed AI Workflows
01:01:28 - How Slack-Based Agent Work Changed Behavior
01:04:03 - Internal Tools, Third-Party Models, and Integration
01:15:21 - How AI Is Changing the Experience of Coding
01:19:54 - Learning, Verification, and Understanding AI-Generated Code
01:25:00 - AI Tools for Better Code Review
01:30:09 - Modular LLM Code Review Tools
01:32:34 - Visualizations and New Interfaces for Understanding Code
01:36:39 - Agent Teams, Testing, and the Future of Software Development
01:45:10 - AI, slop, and what’s worth building
01:46:15 - AI agents and the changing code review contract
01:47:16 - Treating code as a prototype and discussing its merits