Sabitlenmiş Tweet
Letta
738 posts
Letta
@Letta_AI
Building machines that learn. Try Letta Code: the memory-first agent.
npm i -g @letta-ai/letta-code Katılım Ağustos 2024
0 Takip Edilen9.2K Takipçiler
Letta retweetledi
Custom agents let you scope responsibilities in ways that make sense for your org (like with real employees)
At @Letta_AI we have 4 AI employees in Slack with overlapping tools, but different responsibilities:
- Ezra: Customer support + issue creation
- Overload: Manage issue->PR pipeline (including implementation)
- Noam: Manage training/eval jobs
- Brad: Office management + ops
David Cramer@zeeg
vendor-specific chatbots are broken by design that means the Sentry agent, the Linear agent, and any others you might have in Slack they are fine for some point situations, they're nice to get started with, but agents with generalized access outperform them in every single scenario some weeks ago we built an internal Slackbot, gave it access to a bunch of systems (Sentry, GitHub, Linear, Notion, etc), and its capabilities overnight far exceed these other bots "Oh cool Linear can now search your code bases" - our bot did that on day one, and then could push that information wherever it needed to go. Its useful to the point where I now discourage use of things like the Linear bot because it _creates worse outcomes_. this also goes beyond the simple generalization of access: we can customize it. we throw in skills-as-runbooks, templates, etc and the outcomes once again incrementally improve if your org hasnt already built a general purpose bot internally you should. if you need inspiration ours is open source on GitHub (albeit fairly unstable still) github.com/getsentry/juni…
English
The recording of Letta's office hour with @just_cameron for May 14th, 2026 is now available.
This week focused on the next Letta Code release: a more memory-forward UI, local mode, slash-invoked skills, /goal for long-running objectives, and changes to our plans.
Table of contents
00:00 Welcome
00:54 New memory-forward UI
01:16 Built-in markdown memory editor
03:27 Memory graph and commit history
04:05 Agent profile cards and avatars
05:08 Local mode: no Cloud, no Docker
07:03 Slash commands for skills
07:22 /goal for long-running objectives
08:25 Max and Max Lite are being sunset
11:08 Pro, BYOK, credits, and Codex
12:51 Q&A begins
English
Letta's office hours livestream is today at 11:30am Pacific.
Watch @just_cameron present the latest in the Letta ecosystem and field questions from users.
Link below 👇
English
Letta Code deep dive: work with agents that learn
0:00 – Intro
0:58 – Letta Code is agent-centric
1:34 – Viewing your agent's memory (MemFS)
2:22 – Memory graph view
4:01 – Working in parallel with conversations
5:40 – Remote environments
6:46 – Forking
10:35 – Dreaming (sleep-time)
12:20 – Letta Code's subagents
14:16 – Connecting BYOK / coding plans
15:05 – Memory initialization (/init)
16:17 – Memory Doctor (/doctor)
17:03 – Skills
17:47 – Schedules (heartbeats) and Channels
18:26 – Outro
English
The recording for today's office hours is now available on YouTube.
This week:
- Our Slack integration
- Digital coworkers
- Remote environments
- Letta is not a memory layer
See the preview in this post, or the full video below 👇
Preview chapters:
0:00 Intro
0:35 Slack channels recap
1:40 Meet Brad, the office manager agent
7:29 Experiential AI: correcting Brad with a screenshot
8:34 Other agents on Letta's Slack (Sensemaker, Loop)
9:13 Reference to the digital coworker walkthrough video
12:30 Remote environments and Railway one-click deploy
13:44 Per-environment channel and schedule config
14:35 Kimi K 2.6 first impressions
15:47 Q&A begins
English
Letta retweetledi
When messages to a Letta Code agent go through a channel (e.g. Slack), you can see the source + name rendered in the app!
English
Learn how to add a digital coworker to your slack
0:00:00 – Intro: what's a digital coworker?
0:01:54 – Spinning up your Letta agent
0:03:05 – Creating the Slack app (tokens + scopes)
0:06:07 – Plugging Slack into Letta Code
0:09:33 – Live demo: lookups, vision, memory
0:12:41 – Letting your agent handle GitHub PRs
0:16:35 – Skills & scheduled tasks
0:21:31 – Running your own Letta server
English
Watch this video to learn how to build your own digital coworker with Letta Code and Slack: youtu.be/3GqHXjCD7qA
YouTube
English