--- schema: 1 type: video title: 'Brains vs Hands: How to Run AI Agents Safely in Production — Viren Baraiya' description: >- Let the model decide what should happen next. Never let it improvise how a production cluster gets restarted. Viren Baraiya, co-founder and CTO of Orkes and the original creator of Netflix Conductor, explains how to run agents in production. Real agents aren't just chatbots: they run on schedules, react to events, coordinate other agents and can run for months, so the harness around them starts to look like an application, much like a set of microservices. His key idea is a clean split between the brain and the hands. The LLM plans what should happen next; a deterministic, durable harness does the actual work, with approval gates, idempotent steps and recorded side effects. He frames agent harnesses as late-bound sagas, workflows the agent assembles at runtime, and demos an SRE agent whose plan is compiled into a Conductor workflow, executed, checked and re-planned over two loops. In this talk: • Background, scheduled, event-driven and long-running agents • Why the harness is the application, mixing deterministic and non-deterministic parts • The brain and the hands: the LLM plans, the harness executes • Demo: an SRE agent's plan compiled into a durable workflow SPEAKER Viren Baraiya, Co-founder & CTO, Orkes X: https://x.com/virenbaraiya LINKS Orkes: https://orkes.io Orkes on X: https://x.com/orkesio CHAPTERS 0:00 Intro 0:47 Who runs agents in production? 1:12 Beyond the hello-world agent 1:52 Background, scheduled and event-driven agents 2:27 Coordinators and multi-agent systems 3:02 Agents are applications, not components 3:12 Lessons from microservices 4:22 The harness is the application 5:02 Deterministic and non-deterministic parts 6:02 Harnesses run for seconds or months 6:41 Durability is table stakes 7:26 How the loop works 8:21 The agent plans, the harness executes 9:21 Human approval gates 9:41 Idempotency and side effects 9:56 The brain and the hands 10:21 Agent harnesses as late-bound sagas 11:21 Every possible branch 11:51 Demo: an SRE agent 13:31 Compiling the plan into a workflow 14:16 Demo: the second loop 15:56 Conductor, open source Recorded at the AI Engineer World's Fair 2026 in San Francisco. Subscribe for more talks from the engineers building with AI. AI Engineer: https://ai.engineer YouTube: https://www.youtube.com/@aiDotEngineer X: https://x.com/aiDotEngineer LinkedIn: https://www.linkedin.com/company/aidotengineer/ #AIAgents #WorkflowOrchestration #AIEngineer canonical: >- https://liquidgold.cc/channel/aidotengineer/explore/video/brains-vs-hands-how-to-run-ai-agents-safely-in-prod-b263f4821515 markdown: >- https://liquidgold.cc/channel/aidotengineer/video/brains-vs-hands-how-to-run-ai-agents-safely-in-prod-b263f4821515.md llms: null parent: 'https://liquidgold.cc/channel/aidotengineer.md' channel: aidotengineer updated: '2026-10-23T22:19:32.226Z' language: null children: [] videoId: NaOkR3VSfR4 durationSeconds: 1010 publishedAt: '2026-10-08T22:30:09.000Z' searchChunks: - ordinal: 0 text: null startSeconds: 90 endSeconds: 112 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=90' heldReason: rights - ordinal: 1 text: null startSeconds: 112 endSeconds: 123 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=112' heldReason: rights - ordinal: 2 text: null startSeconds: 123 endSeconds: 158 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=123' heldReason: rights - ordinal: 3 text: null startSeconds: 158 endSeconds: 169 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=158' heldReason: rights - ordinal: 4 text: null startSeconds: 169 endSeconds: 192 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=169' heldReason: rights - ordinal: 5 text: null startSeconds: 203 endSeconds: 271 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=203' heldReason: rights - ordinal: 6 text: null startSeconds: 271 endSeconds: 283 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=271' heldReason: rights - ordinal: 7 text: null startSeconds: 283 endSeconds: 307 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=283' heldReason: rights - ordinal: 8 text: null startSeconds: 307 endSeconds: 318 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=307' heldReason: rights - ordinal: 9 text: null startSeconds: 318 endSeconds: 363 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=318' heldReason: rights - ordinal: 10 text: null startSeconds: 373 endSeconds: 407 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=373' heldReason: rights - ordinal: 11 text: null startSeconds: 407 endSeconds: 418 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=407' heldReason: rights - ordinal: 12 text: null startSeconds: 464 endSeconds: 510 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=464' heldReason: rights - ordinal: 13 text: null startSeconds: 523 endSeconds: 567 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=523' heldReason: rights - ordinal: 14 text: null startSeconds: 567 endSeconds: 578 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=567' heldReason: rights - ordinal: 15 text: null startSeconds: 578 endSeconds: 588 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=578' heldReason: rights - ordinal: 16 text: null startSeconds: 599 endSeconds: 610 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=599' heldReason: rights - ordinal: 17 text: null startSeconds: 632 endSeconds: 644 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=632' heldReason: rights - ordinal: 18 text: null startSeconds: 689 endSeconds: 700 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=689' heldReason: rights - ordinal: 19 text: null startSeconds: 722 endSeconds: 820 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=722' heldReason: rights - ordinal: 20 text: null startSeconds: 820 endSeconds: 831 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=820' heldReason: rights - ordinal: 21 text: null startSeconds: 866 endSeconds: 876 watch: 'https://liquidgold.cc/watch/NaOkR3VSfR4?t=866' heldReason: rights --- # Brains vs Hands: How to Run AI Agents Safely in Production — Viren Baraiya Let the model decide what should happen next. Never let it improvise how a production cluster gets restarted. Viren Baraiya, co-founder and CTO of Orkes and the original creator of Netflix Conductor, explains how to run agents in production. Real agents aren't just chatbots: they run on schedules, react to events, coordinate other agents and can run for months, so the harness around them starts to look like an application, much like a set of microservices. His key idea is a clean split between the brain and the hands. The LLM plans what should happen next; a deterministic, durable harness does the actual work, with approval gates, idempotent steps and recorded side effects. He frames agent harnesses as late-bound sagas, workflows the agent assembles at runtime, and demos an SRE agent whose plan is compiled into a Conductor workflow, executed, checked and re-planned over two loops. In this talk: • Background, scheduled, event-driven and long-running agents • Why the harness is the application, mixing deterministic and non-deterministic parts • The brain and the hands: the LLM plans, the harness executes • Demo: an SRE agent's plan compiled into a durable workflow SPEAKER Viren Baraiya, Co-founder & CTO, Orkes X: https://x.com/virenbaraiya LINKS Orkes: https://orkes.io Orkes on X: https://x.com/orkesio CHAPTERS 0:00 Intro 0:47 Who runs agents in production? 1:12 Beyond the hello-world agent 1:52 Background, scheduled and event-driven agents 2:27 Coordinators and multi-agent systems 3:02 Agents are applications, not components 3:12 Lessons from microservices 4:22 The harness is the application 5:02 Deterministic and non-deterministic parts 6:02 Harnesses run for seconds or months 6:41 Durability is table stakes 7:26 How the loop works 8:21 The agent plans, the harness executes 9:21 Human approval gates 9:41 Idempotency and side effects 9:56 The brain and the hands 10:21 Agent harnesses as late-bound sagas 11:21 Every possible branch 11:51 Demo: an SRE agent 13:31 Compiling the plan into a workflow 14:16 Demo: the second loop 15:56 Conductor, open source Recorded at the AI Engineer World's Fair 2026 in San Francisco. Subscribe for more talks from the engineers building with AI. AI Engineer: https://ai.engineer YouTube: https://www.youtube.com/@aiDotEngineer X: https://x.com/aiDotEngineer LinkedIn: https://www.linkedin.com/company/aidotengineer/ #AIAgents #WorkflowOrchestration #AIEngineer Canonical: https://liquidgold.cc/channel/aidotengineer/explore/video/brains-vs-hands-how-to-run-ai-agents-safely-in-prod-b263f4821515 · Channel: [AI Engineer](https://liquidgold.cc/channel/aidotengineer.md) > Search chunk text is held until public-source provenance is verified. ## Document chunks --- #### Chunk 0: Document excerpt (0:12 to 1:18) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 1: Document excerpt (1:18 to 3:12) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 2: Document excerpt (3:12 to 5:07) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 3: Document excerpt (5:07 to 6:03) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 4: Document excerpt (6:03 to 7:32) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 5: Document excerpt (7:32 to 8:30) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 6: Document excerpt (8:30 to 9:27) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 7: Document excerpt (9:27 to 10:32) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 8: Document excerpt (10:32 to 11:51) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 9: Document excerpt (11:51 to 13:51) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 10: Document excerpt (13:51 to 16:03) > Held: transcript rights do not permit this chunk's text. --- #### Chunk 11: Document excerpt (16:03 to 16:50) > Held: transcript rights do not permit this chunk's text.