The Agentic Infrastructure Gap: In-Distribution Languages Make It a Coding Problem | Joe Duffy @scaleconference
The Agentic Infrastructure Gap: In-Distribution Languages Make It a Coding Problem | Joe Duffy  @scaleconference
Uploaded June 2026 | Updated September 2026, 1 week ago
Speaker: Joe Duffy from Pulumi

To handle the scale and velocity of AI-written code, we will have no choice but to let AI manage our infrastructure too. Yet Andrej Karpathy recently described getting an app running in production as “assembling IKEA furniture”: cloud consoles, API keys, copy-pasted config, glue, ... all things that sit outside the code an LLM can reason about. Frontier models are trained on billions of lines of real languages like Python, TypeScript, and Go, and vanishingly little bespoke DSLs and manual procedures. By modeling infrastructure in code space so the LLM can do what it does best — code — we just need an oracle that can map code changes back to infrastructure outcomes. In this talk, I’ll share what we’ve learned at Pulumi working alongside leading AI companies and frontier labs to build for a world where agents manage infrastructure. The platforms that win in this new era will look different, but many of the human-ergonomic benefits of programming languages are what will get us there.

Learn more about the @Scale conferences here: atscaleconference.com
The Agentic Infrastructure Gap: In-Distribution Languages Make It a Coding Problem | Joe DuffyLightning Talk: AI-Native Network Capacity Management | Mohab Gawish, MetaLive from SCCC: MetaRoCE: From Spec to NIC to Open Source | Balakrishnan Raman and Sandeep NagarajEvolving GenAI Media Infrastructure Deployments | Rushaan Mahajan, Sima LabsOur Journey to Safely Unleash Agents at Meta Scale | David Pariag from MetaLive from SCCC: MetaRoCE: Meta’s RDMA Transport | Arvind Srinivasan and Kingshuk MandalInside Instagrams VR Revolution: AI-Powered 3D Content at Scale🚀 Registration is LIVE!Co-Designing Communication for AI Accelerators | Rajeev Nair and Wes Bland from MetaTrack 1 - Live Q&A Session #2 by Shashi GandhamMetaRoCE: Meta’s RDMA Transport | Arvind Srinivasan, Meta and Kingshuk Mandal, KeysightEnhancing Runtime Reliability in LLM Training via Fine-Grained Observability - Live from SCC
@Scale |

The Agentic Infrastructure Gap: In-Distribution Languages Make It a Coding Problem | Joe Duffy

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER