Letta Office Hours is Letta's weekly livestream about building and using persistent agents. Cameron usually opens with recent product changes and a working demo, sometimes brings in a Letta team member to explain a product surface or decision, and then leaves most of the room for audience questions. Episodes move freely between release notes, technical architecture, community projects, failure reports, and the practical experience of keeping one agent over time.
Recordings are published on YouTube.
The format
Office Hours has a recurring shape without becoming a rehearsed webcast:
- What changed. A short roundup of releases, model support, experiments, and current product work.
- Show the thing. A live demonstration of a Letta feature, an agent workflow, or a community project.
- Guest conversation. When a team member joins, Cameron helps make their product area, design decision, or open problem legible to the audience.
- Open Q&A. The longest part of many episodes, where viewer questions lead into debugging, architecture, product philosophy, and concrete use cases.
The balance changes from week to week. Some episodes are focused release tours; others become long conversations about architecture, product judgment, or how people use their agents. The loose format is part of the point: questions can expose confusion, missing capabilities, and design tradeoffs that a prepared launch video would skip.
About these guides
The original recording remains primary. Each guide adds a compact lead, selected timestamp links, explanatory sections, related public sources, and an architectural synthesis so the episode remains useful without reproducing its transcript.
The pages are historical records, not current product manuals. Letta changes quickly; a feature described in an older episode may have been renamed, redesigned, removed, or absorbed into a different surface. Current documentation remains the source for current behavior.
Episodes
2026
2026-08-13: Dynamic Workflows, Cloud Agents, Grok 4.6, and Personalized Software — Agent SDK examples, dynamic workflows, cloud sandboxes, MemFS retrieval, application architecture, and software built around persistent agents.
2026-08-06: Shared Memory, Hypervigilant, Schedules, and the Agent SDK — Shared memory, schedule execution, local providers, the channels gateway, Hypervigilant, and the Agent SDK.
2026-07-30: Free Dreaming, ACP, MCP, and Trajectory — Letta adds free dreaming on Cloud, ACP editor support, broader MCP integration, and Trajectory for learning across coding-agent harnesses.
2026-07-23: Agent SDK, Shared Memory, and Spec-Driven Development — July 23, 2026 office hours episode on the Letta Agent SDK, agent-created schedules, shared memory, ACP support, and spec-driven development.
2026-07-16: Cloud Sandboxes, GitHub Integration, GPT-5.6, and Agent Sharing — Letta adds persistent cloud sandboxes, GitHub integration, model support updates, and a broader discussion of shared and cross-organizational agents.
2026-07-09: GPT-5.6, Grok 4.5, Letta App Server, and Spec-Driven Agents — Office hours cover new Slack agent UI, GPT-5.6 and Grok 4.5, the new-user path in Letta Chat, and the app server and Agent SDK architecture.
2026-07-02: Letta Agent, Mod Challenge Demos, Slack Agents, and GLM-5.2 — Cameron frames Letta Agent as the broader packaged experience, then walks through mod challenge demos, Slack coworker workflows, model choices, and subagent tooling.
2026-06-29: Mods, Signal Support, GLM-5.2, and Slack Agents — Office hours on mods, Signal channels, model routing, secret handling, and why Slack-style channels can make agents feel more like coworkers.
2026-06-19: Mods, MemFS, and Teaching Agents New Tricks — Office hours on mods, local mode, channels, schedules, memory architecture, and how Letta teaches agents new capabilities over time.
2026-05-21: New Desktop UI, Local Mode, Agent Services, and Skills vs MCP — Office hours on the redesigned Letta Code desktop app, local mode, memory UX, agent services, and the product and protocol tradeoffs behind skills, MCP, and channels.
2026-05-14: Local Mode and the New Pro Plan — Office hours on a memory-forward Letta Code UI, local mode, slash commands, /goal, and the shift from Max plans to Pro, BYOK, and credits.
2026-05-08: Sandboxes for Letta Chat, Sensemaker, and an agent that plays Pokemon — New channel controls, local-only Letta Code, sandboxed agents in Letta Chat, Sensemaker, and design guidance for building durable agents.
2026-05-01: Discord Channel, Custom Plugins, Letta City Sim, and a Special Guest — Office hours covers Discord channels, custom plugins, schedules, policy boundaries, Letta City Sim, model updates, and a long Q&A on memory, context, and agents.
2026-04-23: Digital Coworkers with Slack — Letta office hours on Slack-native agents, remote environments, channel hosting, and why skills beat MCP-style integrations.
2026-04-16: Channels, Cron, and Opus 4.7 — Office hours on built-in channels, schedules, remote environments, and how recent model and agent changes shape always-on workflows.
2026-04-09: April 9th, 2026 — Office hours covering Letta Code general access, remote control mode, the memory viewer, social agents, ChatGPT memory import, and the Context Constitution.
2026-04-02: April 2nd, 2026 — Cameron introduces the new Letta Code app, demos memory repair and sleep-time processing, and answers questions on tools, deployment, and agent memory design.
2026-03-26: Desktop Early Access, Forking, Secrets, Memory Doctor — Office hours covering desktop early access, new slash commands, memory health checks, auto fallback, and the shift toward stateful multi-agent workflows.
2026-03-19: March 19th, 2026 — Cameron explains Letta’s client-side shift, MemFS, sleeper agents, deprecations, and the move toward skills, code SDK orchestration, and computer-use-first workflows.
2026-03-12: March 12, 2026 — Office hours recap of Auto Mode, Letta Code and Chat changes, Lettabot improvements, Ezra's new capabilities, and community-built fleet and team tooling.
2026-03-05: Letta Remote, Claude Subconscious Demo, and Lettabot — Office hours on Letta Remote, chat-to-remote workflows, Lettabot updates, pricing tradeoffs, and demos of Claude Subconscious and Co-work.
2026-02-26: MemFS, Letta Chat, and the future of AI agent memory — Office hours on MemFS, Letta Remote, Letta Chat, sleep-time compute, Lettabot upgrades, and how Letta is rethinking agent memory.
2026-02-05: Opus 4.6, Lettabot Updates, Agent File Directory, and More — Office hours on Opus 4.6, Letta Code provider setup, Lettabot, agent self-forking, and why Letta favors stateful agents over retrieval-only memory.
2026-01-29: Introducing LettaBot +Claude Subconscious Demo — Office hours on LettaBot, a locally deployed messaging bridge for agents, plus a live Claude Subconscious demo that injects Letta context into Claude Code.
2026-01-26: Letta Chat, GitHub Action, Letta Code, and more! — Cameron demos Letta Chat, a GitHub Action for Letta Code, specialist repo agents, the note tool, and social agents on Blue Sky.
2026-01-09: Message Scheduling, GitHub Actions, Ralph Mode, and the Note Tool — Office hours on scheduling, GitHub Actions, Ralph mode, custom commands, sub-agents, the note tool, and why simple memory structures often beat heavier abstractions.
2025
2025-12-18: Letta Code Demo, Agent Skills, Claude Code Proxy & More — Office hours on Letta Code, agent skills, persistent memory, sub-agents, and emerging ideas for shared blocks and file-system-like agent environments.
2025-12-04: December 4th, 2025 — December 4, 2025 office hours on the v1 SDK launch, message search, Learning SDK examples, skills, lettactl, personal agents, and Ezra updates.
2025-11-06: November 6th, 2025 — November 6, 2025 office hours on the v1 SDK migration, shared archives, Letta Code improvements, AI Memory SDK v0.2, scheduling, and Ezra.
2025-10-30: October 30th, 2025 — October 30, 2025 office hours on Letta Code, Co, the Obsidian plugin, personal agents, and the platform shift toward terminal-first and note-centric workflows.
2025-10-16: October 16th, 2025 — October 16, 2025 office hours on voice support, the runs viewer, Telegram improvements, stateful-agent patterns, and practical tips for building with memory.
2025-10-02: October 2nd, 2025 — October 2, 2025 office hours on the v1 alpha, memory tools, mobile agents, Obsidian, Bluesky, and the shift toward a new agent architecture.
2025-09-25: September 25th, 2025 — Office hours on the AI SDK v5 update, cloud reliability work, open-source release planning, memory behavior, and model/provider recommendations.
Recurring themes
Persistent identity and memory. The series repeatedly separates an agent from any one chat window, model call, or client application.
Execution surfaces. Letta Code, Desktop, Chat, channels, sandboxes, schedules, and the Agent SDK appear as different ways to connect durable agents to work.
Harness extension. Skills, tools, MCP, mods, protocols, and provider compatibility are treated as different layers rather than synonyms.
Inspectable state. Memory files, commits, runs, approvals, traces, and tests make agent behavior reviewable instead of merely fluent.
Teaching and evaluation. Later episodes increasingly ask how agents learn from trajectories, specifications, feedback, and explicit release gates.