Ship great agents fast with our open source frameworks – LangChain, LangGraph, and Deep Agents. Maintained by @LangChain.
Joined January 2026
- Tweets 258
- Following 5
- Followers 3,364
- Likes 100
85 Photos and videos
Jun 12
We'd love your feedback!
Jun 12
we revamped our skills docs this week!
what are they missing? what questions do you have about skills that are unanswered here?
docs.langchain.com/oss/pytho…
2
5
4,342
LangChain OSS retweeted
Jun 12
everybody's talking about loops!! how can you instrument them with langchain?
1. token loop supported by a model (choose any model with langchain)
2. create_agent gives you the agent loop (model tools repeat until done)
3. deepagents gives you the self verification loop (agent loop verify repeat until satisfied)
4. deployments give you the meta loop (trigger agent runs in reaction to events that help improve a system)
5. i think the ??? loop is what we're trying to close with engine: run an agent over each trace and figure out what to tweak - prompts, tools, self verification, etc so that your meta loop is more effective per cycle.
Jun 12
[AINews] Loopcraft: The Art of Stacking Loops
@RichardSSutton has his “Bitter Lesson” for models. We now have the Salty Lesson for agents:
Don’t fix things yourself, as you have done historically.
Instead focus on systems that scale with more agents, like goals and orchestration.
More in today's op-ed: latent.space/p/ainews-loopcr…
7
23
94
28,564
Jun 8
Rubrics are even more flexible than /goal
You can define a custom subagent for the grading, including custom tools, prompt, and iteration limits.
Try it out and let us know what you think!
Jun 8
we just shipped support for rubrics in deepagents ✅
give your agent a clear definition of what "done" looks like, and force it to run in a loop until said goal is complete
this is similar to /goal in claude code, but works for any agent (not just a coding agent)
1
7
2,376
Jun 4
One of the most common questions we get from agent builders is how to make agents resilient to failures, so we wrote a guide!
1
6
1,172
LangChain OSS retweeted
Jun 4
let's assume agent = model harness
unfortunately, good models are getting really expensive!
so you need a great harness to compensate.
you can close the gap on agent performance by hill climbing w/ your harness, here's a guide on how to do just that!
langchain.com/blog/tuning-de…
3
13
83
15,545
Jun 4
open models 🤝 open harness
Jun 4
transitions like this are why we think it's helpful to have a provider-agnostic harness
we used to talk more about swapping models when the latest and greatest came out -- but the latest and greatest from major providers are expensive!
we're seeing more folks swap to open source models for cost reasons, which is easy with deepagents and/or langchain
1
1
9
4,098
Jun 3
LangChain makes it easy to build a bespoke harness for your use case.
4
361
LangChain OSS retweeted
Jun 3
it's about to be the end of an era
7
8
58
6,632
Jun 2
RubricMiddleware helps your agent verify task completion with a grader subagent
This is similar to /goal in claude code or codex, but for deepagents!
1
11
2,364
Jun 2
The quarq agent is built on LangGraph!
LangGraph makes it easy to build complex memory systems (quarq is now at the top of the LongMemEval leaderboard)!
1
6
16
3,473
LangChain OSS retweeted
Jun 1
here's a quick overview of
a) what is deepagents
b) what makes deepagents good at complex tasks
c) how to easily take one to production!
youtube.com/watch?v=LdQpoK2T…
2
15
80
19,404
May 26
create_agent is the easiest way to get started building an agent! it provides an incredibly flexible interface for building agents designed around your use case!
May 26
we just revamped the create_agent docs!
the new agents page shows how to build a custom harness for your use case w/ create_agent as an easy entrypoint
start w/ your prompt and tools, then add middleware to customize at any point in the loop
docs.langchain.com/oss/pytho…
2
1
9
3,001
LangChain OSS retweeted
May 20
2
23
113
43,834
May 20
You can now add an interpreter to deep agents with two lines of code. This enables some capabilities that were either locked behind provider APIs, or too hard to adopt in a meaningful way.
Go read this write up for more info on what this is and how to use it!
May 20
we're adding interpreters to our agents, and now so can you!
There's a lot of interesting patterns/ agent behaviors I've seen come across the timeline that depend on some way to run code (think PTC, CodeMode, RLM), but it's been hard to filter that down into an abstraction thats easy to work with. We've spent the last couple of weeks creating an abstraction to solve just that!
Expect me to yap more about this soon - there's a lot to talk about here that was hard to fit into a single post
1
3
14
5,551
May 19
We're adding new community middlewares left and right!
Let us know about your favorite middlewares and we'd love to feature them!
docs.langchain.com/oss/pytho…
May 19
start your tool execution mid-model call with this new eager-tools middleware!
tools execute as soon as their block finishes streaming, reducing end to end latency!
s/o @bmd1905 for this awesome new middleware!
source: github.com/cloudthinker-ai/e…
docs: docs.langchain.com/oss/pytho…
2
2
5
1,047
May 18
ICYMI: we shipped Deep Agents v0.6 last week, our biggest release yet!
3
3
10
2,541
May 13
✅ Harness profiles: Per-model tuning support for open models (@Kimi_Moonshot, @Alibaba_Qwen @deepseek_ai)
✅ Code interpreter: A programmable runtime inside the agent loop
✅ Streaming-typed projections for messages, tool calls, subagent events
✅ DeltaChannel: Efficient checkpoint storage for agents
✅ ContextHubBackend: Store skills, policies, memories that shape agent behavior
💻 langchain.com/blog/deep-agen…
4
13
101
19,627
LangChain OSS retweeted
May 12
1
10
61
28,096
May 11
Swap models & view their capabilities! Try out in Deep Agents CLI:
docs.langchain.com/oss/pytho…
Mar 7
here's model profile details look like in practice, using @NVIDIAAIDev's Nemotron models as an example:
3
5
20
4,169