Sunil Pai created Partykit, the open source tool for real-time multi-player apps. For his day job, he builds AI Agents at Cloudflare.

Speaker info:
- https://sunilpai.dev/
- blog.cloudflare.com/author/sunil/
- linkedin.com/in/sunil-pai-a47732253/

Timestamps
0:00 Introduction and speaker background
1:16 What is "Code Mode"?
1:31 Limitations of traditional tool calling at scale
2:03 The shift to generating executable code
3:01 Scaling API usage at Cloudflare
4:05 Why code generation is more efficient
5:28 Live demonstration of the Mythical server
7:20 A new way of interacting with systems
9:09 Example: The Kenton canvas and tic-tac-toe anecdote
11:46 New software architecture: The "Harness"
13:28 Observability and security in sandboxed environments
14:15 Long-running workflows and generative UI
16:41 Future outlook: Building for the next generation of users
17:50 The resurgence of capability-based security
18:33 Conclusion and final thoughts

AI Engineer

Sunil Pai from Cloudflare presents 'code mode', an approach where LLMs generate executable code (JavaScript) instead of making repeated JSON tool calls. This dramatically reduces token usage — shrinking Cloudflare's 2,600-endpoint API surface from ~1.2 million tokens to ~1,000 tokens using just two tool calls (search and execute). The talk explores a new software architecture pattern: a capability-based sandbox that starts with zero permissions and grants APIs explicitly, enabling safe code execution with full observability. Pai argues this unlocks generative UIs, long-running workflows, and a new paradigm where AI agents interact with systems through code rather than predefined tool schemas — effectively democratizing programmatic system access beyond developers.

Code Mode - Sunil Pai, Cloudflare