Dont Vibe Code Your Agent Infrastructure - Akshat Bubna, Modal @aiDotEngineer
Dont Vibe Code Your Agent Infrastructure - Akshat Bubna, Modal  @aiDotEngineer
Uploaded November 2026 | Updated November 2026, 5 days ago
Akshat Bubna, CTO of Modal, explains what AI agents actually need from the infrastructure underneath them, starting with a spreadsheet where every cell is backed by its own agent. He walks through the building blocks: a workspace for each agent, state that persists for days without paying for idle compute, resources that stretch and shrink as the agent works, GPUs on demand, shared file systems for agents that collaborate, network controls that keep credentials and data from leaking, and observability agents can read themselves. He makes the case against vibe coding your own sandbox platform, shares what scale looks like today, and describes where agent infrastructure is heading.

Chapters:
0:00 Intro
0:24 Every spreadsheet cell backed by an agent
1:33 What that architecture needs
3:15 What is Modal?
4:03 Agents need a workspace
5:03 Workspaces that persist
6:21 Agents need elastic resources
7:41 Agents need GPUs
9:02 Agents collaborating through shared files
9:57 Keeping agents from leaking data
11:20 Observability for agents
12:12 Why not to vibe code a sandbox platform
13:31 What scale looks like
14:54 Separating the brain from the hands
15:37 From pets to cattle to microbes

Akshat Bubna
modal.com

Join us in SF for the second annual AIE CODE summit: https://aie.codes our last flagship summit of 2026. Applications live now!
Dont Vibe Code Your Agent Infrastructure - Akshat Bubna, ModalHow to Know When an AI Agent Can Work Unsupervised — Evan Kormos, CoinbaseYou Cant Tell What an Agent Skill Will Do by Reading It - Roey Ben Chaim, ZenityGive Your On-Call Its Own AI Teammate - Swaapnika Motapothula, Chinmay Gandhi & Vlad Tsvang, MetaWhy AI Writing Still Reads Like Slop — Louis-François Bouchard, Towards AIHow a Trading Firm Ships New AI Models the Same Day - Noyan Tokgozoglu, OptiverWhy Frontier Models Fail as Buy-Side Analysts - Janie Gu, Google DeepMind & Suyeol Yun, LinqAlphaA Bank Built Its Own Coding Agent. It Took Off — Felippe Felisola Caso & Francesco Galletta, MonzoAI Engineer NYC Livestream | Bridgewater, Two Sigma, Arize, Coatue, Nous (Hermes), Gen. IntuitionDoes Code Quality Still Matter? It Depends on This — Max Kanat-Alexander, Capital OneStop Shipping LLM Settings That Only Win on Paper - Rohan Pandey & Balaji Varadarajan, DigitalOceanHow to Test AI Agents When No Benchmark Exists — Nathanael Weill, Air Canada
AI Engineer |

Don't Vibe Code Your Agent Infrastructure - Akshat Bubna, Modal

SHARE TO X SHARE TO REDDIT SHARE TO FACEBOOK WALLPAPER