Harrison Chase(@hwchase17)

building evals is hard! we're working on some skills to try to automate as much as possible. still ...

Harrison Chase (@hwchase17) — Wed, 22 Jul 2026 19:28:29 GMT

building evals is hard!  we're working on some skills to try to automate as much as possible. still requires human in the loop, but should help bootstrap

overall flow is:
- give coding agent the codebase + actual traces
- iterate on eval direction with user
- build evals (using harbor)
- run evals
- look at results, iterate with user again, repeat

Viv@Vtrivedy10

x.com/i/article/2079…

💬20🔄19❤️185👀20883📊54

native integrations for four leading voice frameworks so you can see whats happening dont fly blind

Harrison Chase (@hwchase17) — Tue, 21 Jul 2026 17:30:41 GMT

native integrations for four leading voice frameworks so you can see whats happening

dont fly blind

LangChain@LangChain

Voice agents are exploding. Don’t let them be a black box in production.

Today, we’re launching LangSmith tracing for 4 voice frameworks:
🎙️ @pipecat_ai
🎙️ @livekit
🎙️ @OpenAI Realtime
🎙️ @GeminiApp Live (Google ADK)

Learn molangchain.com/blog/trace-voi…Fkmj

💬10🔄7❤️32👀4080📊11

great post by jeff! a few thoughts I had as I read along, wrt what we're building at langchain: > d...

Harrison Chase (@hwchase17) — Tue, 21 Jul 2026 02:33:20 GMT

great post by jeff! a few thoughts I had as I read along, wrt what we're building at langchain:

> design a compound learning machine

as you use or deploy agents, they need to be learning! high level, i believe traces are the core of this - you need to see what happened, get feedback on it, and then update the system

> own your context, rent your intelligence

agree with this sentiment - own your "tacit and institutional knowledge", make it easy to swap out models. i think the phrasing will get confusing tho - what exactly is intelligence? @satyanadella had a great post here (x.com/satyanadella/s…) where he says a lot of things that agrees with this, but he also says stuff like "In consuming intelligence, you are creating intelligence... This is your particular intelligence". the nuance is there is general intelligence (eg the models, which you should rent) and then your company proprietary intelligence, which you should own. both could reasonably be called intelligence though, so its confusing

> build your tools, buy infra

buy the lowest common denominator infra, build specific things on top of it. at LC we try to focus on lowest common denominator infra (observability, deployment, sandboxes), but we can do better - we have some cool stuff coming in that regard

> develop rubrics

evals are important!

> store and learn from production traces

yup. was this article just an ad for langsmith all along? who knows!

Jeff Huber@jeffreyhuber

x.com/i/article/2079…

💬22🔄15❤️117👀29062📊49

LangSmith Engine is our in product agent that runs over traces, clusters them into issues, and propo...

Harrison Chase (@hwchase17) — Mon, 20 Jul 2026 17:27:19 GMT

LangSmith Engine is our in product agent that runs over traces, clusters them into issues, and proposes fixes

It's a long running, complex, ambiguous process

We wrote about how we evaluate it!

Brace@BraceSproul

x.com/i/article/2079…

💬10🔄8❤️88👀12697📊21

So i didn’t really know what graph engineering is, and i still don’t really… but it’s basically just...

Harrison Chase (@hwchase17) — Mon, 20 Jul 2026 15:00:06 GMT

So i didn’t really know what graph engineering is, and i still don’t really… but it’s basically just langgraph?

Shann³@shannholmberg

what's the difference between a loop and a graph? (marketing edition)

both are ways to run an agent, the difference is who decides the path, the agent or you.

a loop still starts with you. you set the goal, the brief, and the bar it has to clear. what the agent owns is the path.

take writing an SEO article: hand it the brief and it drafts, reads the draft back against that brief, rewrites the weak parts, checks again, and keeps circling until it clears the bar. the one thing you did not write is the step-by-step it took to get there.

a graph is you drawing the steps and the routes between them ahead of time. same article, but now you set the map: research the keyword and the competitors ranking for it. draft from what you find. score that draft against your rubric.

if it clears, add the internal links and publish. if it misses, back to the draft. the agent still decides how to handle each step, it just travels the routes you laid down.

the shape of this has a name, a state machine. every node is a state the work can be in, and a check at each one decides where it goes next, forward when it clears or back to an earlier node when it misses.

if you have built a workflow in n8n, you have already drawn one. nodes wired together, branches that fire on a condition, a step that loops until it clears, that picture is a graph. an agent graph is the same shape, the nodes hold agents doing the work instead of single api calls.

the way I think about it, a graph is a map of loops and checkpoints. some nodes run once, others are their own loop where the agent works something out, and the checkpoints between them read the result and route the work. you keep laying down nodes and checkpoints until the map reliably gives you the output you want.

the vault accelerator I run at my agency is one of these maps, 3 sessions that hand off in a fixed order:

> research session: reads our company brain and past campaign results, pulls in competitor and market context, and builds the cohort we go after
> landing page session: takes that research and builds the page from it
> content session: uses the research and the page to write the copy, illustrations, and slides for the live sessions we run

inside the content session runs a loop, a critic scores each draft against a rubric and sends it back until it clears the bar. that is one node on the map, the checkpoints between the sessions carry the work from one to the next

a graph earns its extra setup on anything you run every week:

> validation gates the work cannot skip
> a fixed set of routes the job can take
> a clear failure point, you see the exact step something broke on

a loop on its own is enough for the work you only do once, where you don't know the path yet, let the agent find it. graphs earn their place on the jobs you repeat, the content pipeline, the SEO and AEO funnel step by step, the vault accelerator

once the map works you reuse it, feed it the next cohort and the whole pipeline runs again

past the loop, the next thing you design is the map it runs inside.

💬64🔄33❤️508👀68583📊148

Dcode, openswe, openswe review, open wiki… All add up to a fully OSS software factory

Harrison Chase (@hwchase17) — Sat, 18 Jul 2026 20:25:37 GMT

Dcode, openswe, openswe review, open wiki…

All add up to a fully OSS software factory

Brace@BraceSproul

x.com/i/article/2078…

💬13🔄4❤️57👀19094📊20

If you like evals, this is a good role for you

Harrison Chase (@hwchase17) — Fri, 17 Jul 2026 21:04:06 GMT

If you like evals, this is a good role for you

Brace@BraceSproul

i need to hire someone who's really really good at evals and benchmarks in SF (or maybe nyc if you're super special)

if this is you, or you know someone who fits this, please dm me! i'd love to chat

💬9🔄3❤️37👀8873📊9

there needs to be an OPEN standard for memory OKF (Open Knowledge Format) is one such standard exc...

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 17:25:52 GMT

there needs to be an OPEN standard for memory

OKF (Open Knowledge Format) is one such standard

excited to announce that we're now using it in OpenWiki

Brace@BraceSproul

x.com/i/article/2077…

💬23🔄52❤️398👀71515📊110

one of the main questions we got about open wiki was whether we would support OKF if you dont know ...

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 17:25:07 GMT

one of the main questions we got about open wiki was whether we would support OKF

if you dont know what OKF is - its a standard for knowledge files

check out Brace's video!

Brace@BraceSproul

I just published a short YouTube video on OKF in OpenWiki. It's only 3:30 long, so if you're curious what OKF is, what it'll mean for OpenWiki, and why we decided to adopt, you should give it a watch!

youtu.be/NxJjMvDN6aE

💬5🔄3❤️31👀7185📊10

fun trip down memory lane

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 17:15:03 GMT

fun trip down memory lane

Jacob Lee@Hacubu

This year @LangChain 🦜 🔗 has almost tripled in size, from ~100 people to over 300! That's a lot of new faces, and I'm often asked: "What made you join early? What was the company like at seven people?"

I'm always honest: I wasn't convinced right away. I mean, three years ago, the website looked like thisweb.archive.org/web/2023030322…BG

Raw HTML and two emojis for a logo. When I told my parents I was joining, they asked, pretty directly, if I'd lost my mind.

I met @ankush_gola11 about a month after ChatGPT launched and the open-source started getting real traction. I was completely fascinated by the technology, but my reaction to his pitch was "cool - good luck!" In addition to said website, I had just come off six years founding a company, and didn't want to jump back into an early startup.

A few weeks later, a client wanted to build an LLM-powered extraction pipeline - nowhere near as trivial then as it is now. I tried an early version of LangChain.js, and it worked great, so I kept exploring and found the section on agents.

The magic there completely hooked me. It truly seemed like something out of science fiction.

Thankfully, @hwchase17 and Ankush were still looking to round out their founding team, and I got to see how they operated up close. Unlike most startups, LangChain has never had a name recognition problem, and I knew firsthand how easy it was to start chasing GitHub stars and other vanity metrics.

But from the start, the team was serious about seeing through the fluff, listening to the community even when tempers flared, and building what actually solved user problems. They weren't always right - I joke that back then we were a six-month-old startup with a year of tech debt - but I admired how quickly they'd throw out popular beliefs and abstractions the moment evidence showed they were wrong or wouldn't scale. Anyone remember ConversationalRetrievalChain?

The website itself was emblematic of this "zero tolerance for bullshit" approach (as Harrison puts it). The small early team could have spent time creating a beautiful marketing site, but marketing wasn't the problem back then (though we're hiring now!). A parrot and a link emoji was enough. Time was better spent shipping.

So, what was it actually like at seven people? Different principles, but ones that rhyme with the old ones. "Zero tolerance for bullshit" has become "act with maximum agency". Our most interesting Slack discussions start with "Hot take", and it's a mark of dishonor if people react with ice cubes, meaning it's not extreme enough to encourage a rethink of a longstanding belief.

But the north star hasn't moved: build something great, and cut everything that isn't that.

There's a lot of story left to write. We're hiring across product, engineering, office management, people ops, and more! If that sounds like a place you'd thrive, come find us.

P.S. the OG site is a lot more clever than it looks - seberkshirehathaway.comyP

💬5🔄1❤️27👀6254📊6

this is one thing ive realized - owning that feedback loop (the signals, the output) is really code ...

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 16:20:46 GMT

this is one thing ive realized - owning that feedback loop (the signals, the output) is really code to owning intelligence

Blum@Blum_OG

@hwchase17 owning the feedback loop is huge

without it youre just hoping it works

💬6🔄2❤️19👀3860📊6

what does it really mean to "own your intelligence"? 1. can the system compound on your terms? th...

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 16:09:22 GMT

what does it really mean to "own your intelligence"?

can the system compound *on your terms*? the durable advantage is not the model/system at this point in time - it's the flywheel that compounds over time

intelligence is no longer just the model - its the system around the model. if the system is opaque or locked away, you are renting intelligence (not owning it)

own the harness. integrate it deeply into your systems. make it specific to the task at hand

own the context and memory layer. includes user preferences, org knowledge, workflow patterns. if memory is trapped in vendor product, your learning is trapped to

own model optionality. you need freedom to route across models, providers, deployment modes. different tasks require different tradeoffs

own the economics. intelligence only matters if you can afford to deploy it broadly. sustainable intelligence at scale

own the feedback loop. observability tells you what happened, evals tell you if it worked. without this loop you cannot measure the intelligence, you are guessing

s/o @nlarusstone@nfcampos for their thoughts on feedback

💬15🔄14❤️84👀12947📊37

dcode with open models! (nemotron 3 ultra)

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 16:00:48 GMT

dcode with open models! (nemotron 3 ultra)

LangChain@LangChain

New video from @its_ao. 

How to run @NVIDIA Nemotron 3 Ultra with @baseten + Deep Agents Code.

✅ 550B parameters, up to 300 tokens per second
✅ Terminal agent w/ skills, sub-agents, MCP support
✅ First-class tracing through LangSmith

Your browser does not support the video tag.

💬3🔄1❤️24👀5994📊5

New Max Agency with @FactoryAI CTO & Co-Founder @EnoReyes. This was a particularly fun one. Since w...

Harrison Chase (@hwchase17) — Thu, 16 Jul 2026 14:36:50 GMT

New Max Agency with @FactoryAI CTO & Co-Founder @EnoReyes.

This was a particularly fun one. Since we're both building harnesses, we nerded out pretty deep.

Awesome conversation on why the harness matters more than the model running underneath it, how Factory built Missions, + more.

⏯️ YouTube: youtube.com/watch?v=HbUznY…

🎧 Apple:podcasts.apple.com/us/podcast/the…Z

🎧 Spotifyopen.spotify.com/episode/5CkX1d…ag

Your browser does not support the video tag.

💬7🔄17❤️92👀34838📊28

Good takes from Caspar on why we’re leaning into slack as an interface for agents

Harrison Chase (@hwchase17) — Wed, 15 Jul 2026 18:19:16 GMT

Good takes from Caspar on why we’re leaning into slack as an interface for agents

Caspar Broekhuizen@caspar_br

x.com/i/article/2077…

💬7🔄9❤️45👀10186📊14

It’s super easy to create custom slack agents now Makes it easy to bring intelligence to where the ...

Harrison Chase (@hwchase17) — Wed, 15 Jul 2026 17:20:21 GMT

It’s super easy to create custom slack agents now

Makes it easy to bring intelligence to where the team already lives

LangChain@LangChain

Fleet in Slack just got a big upgrade.

Now you can bring any Fleet agent into Slack in one click. Give it a custom identity. Use it in channels and threads. Hand off files, approve its actions, and keep all the context of your work in one place.

Read more in our blog: langchain.com/blog/new-in-la…

💬8🔄2❤️43👀11641📊11

Open source coding harness!

Harrison Chase (@hwchase17) — Tue, 14 Jul 2026 20:19:20 GMT

Open source coding harness!

Sure (e/acc)@SureCapital

@LangChain really outdid themselves with dcode. It’s currently one of my favourite harnesses for coding. 🙌🏽

💬9🔄2❤️58👀13038📊14

we did a bunch of work to standardize how coding agents trace to langsmith

Harrison Chase (@hwchase17) — Tue, 14 Jul 2026 18:00:09 GMT

we did a bunch of work to standardize how coding agents trace to langsmith

LangChain@LangChain

Cursor, Copilot, Pi, and OpenCode tracing: Now in LangSmith.

Full session observability, no extra instrumentation.

✅Identify, group, and query any coding-agent trace with the same stable keys, regardless of which agent produced it

✅See the full run tree: turns, model calls, tools, and subagents

✅Token usage and cost per session, out of the box

Blog by @harisaiharishlangchain.com/blog/your-codi…

💬17🔄4❤️67👀14311📊20

> For me, agentic performance on spreadsheet tasks is the whole of Shortcut existence. So I'm goi...

Harrison Chase (@hwchase17) — Mon, 13 Jul 2026 21:42:41 GMT

> For me, agentic performance on spreadsheet tasks is the whole of Shortcut existence. So I'm going to beat competitors by obsessing over the harness.

custom harness is the only way you will beat the labs agentic experience

heres how to build one: https://t.co/Y8QoykQGu5

Peter Wang@BrainsAndTennis

x.com/i/article/2076…

💬14🔄15❤️151👀25913📊52

openwiki is 🚄

Harrison Chase (@hwchase17) — Mon, 13 Jul 2026 21:31:49 GMT

openwiki is 🚄

Brace@BraceSproul

OpenWiki 0.1.2 is now out, with a ton of changes!

Notable updates:
- 🧑‍💻 QOL: `openwiki --init` and `--update` run Code Brain by default (most popular use case!)
- 🔙 Specific prompting to document backlog tasks within wiki files
- 🥳 8 new contributors, with this release including 11 total unique contributors!!

Try it out today, and put up a PR if you have a new feature, fix or improvemengithub.com/langchain-ai/o…VM8A

💬10🔄5❤️40👀7383📊13

if you want to learn more about why coding harnesses are not going to be the optimal harness for sci...

Harrison Chase (@hwchase17) — Mon, 13 Jul 2026 19:14:48 GMT

if you want to learn more about why coding harnesses are not going to be the optimal harness for science: youtube.com/watch?v=RjpTrf…

Nicholas Larus-Stone@nlarusstone

@LatchBio has been on a tear recently. It should be obvious to everyone why coding harnesses are not going to be the optimal harness for science, but it was nice to see someone put numbers to it

💬11🔄6❤️58👀15642📊20

LangSmith does this for you - in the cloud: LangSmith sandboxes & LangSmith deployments - any ...

Harrison Chase (@hwchase17) — Sat, 11 Jul 2026 23:29:58 GMT

LangSmith does this for you 

- in the cloud: LangSmith sandboxes & LangSmith deployments
- any model: LangChain integrates with 100s
- harness: deep agents
- tracing: langsmith observability
- recursive improvement: langsmith engine

Aditya Agarwal@adityaag

Everything in AI moves so fast.

We are going to get here:
- Run all the agents in the cloud
- Choose any model (frontier, OSS, Chinese, American)
- Choose any harness
- Have full tracing
- Have recursive improvement loops

I know it will be there but can it happen already?

💬18🔄13❤️127👀25205📊32

📕LLM Wiki Webinar with @BraceSproul @devstein64 @jeffreyhuber is now on YouTube! Two of my favorit...

Harrison Chase (@hwchase17) — Sat, 11 Jul 2026 19:59:18 GMT

📕LLM Wiki Webinar with @BraceSproul@devstein64@jeffreyhuber is now on YouTube!

Two of my favorite insights from this webinar:

"Wiki as a cache" - @devstein64: basically, the purpose of a wiki is to keep things that are commonly looked up or accessed more top of mind/readily available. What goes in the wiki should be updated and organized with that in mind

"Wiki is a set of hyperlinked pages" - @jeffreyhuber. The evolution of memory (in my view) has gone from:
1. single string
2. file
3. set of files in a directory (wiki)

as you think about scaling up - as files grow, the file structure matters less, and you need links between pages. The internet is a set of hyperlinked pages - so that definitely scales!

Watch it here:youtube.com/watch?v=Lsut4T…U

💬14🔄7❤️83👀11842📊27

great overview of general purpose brain!

Harrison Chase (@hwchase17) — Fri, 10 Jul 2026 17:04:50 GMT

great overview of general purpose brain!

Brace@BraceSproul

I just published a new video on OpenWiki Brains, specifically on the general-purpose brain mode.

In the video I dive into:
- configuring it locally
- its architecture
- what the docs look like
- new features like the 'open questions' agent file 👀

Check it out here:youtu.be/sBg90v2qfasT

💬5🔄7❤️52👀11202📊14

🚀langchain launches this week: all about open source models and memory! First: open source models....

Harrison Chase (@hwchase17) — Fri, 10 Jul 2026 16:39:30 GMT

🚀langchain launches this week: all about open source models and memory!

First: open source models. We partnered with @NVIDIAAI to launch a NemoClaw DeepAgents blueprint. This pairs Deep Agents (our open source, model agnostic harness) with Nemotron 3 ultra (powerful OSS model) and OpenShell (enterprise ready run time). Blog:langchain.com/blog/langchain…L

Second: memory. LLM wikis continue to intrigue us. @BraceSproul@colifran_ released a new version of OpenWiki focused on "personal brains" - creating wikis from gmail, internet, etc:github.com/langchain-ai/o…p

open models and memory pair well together as well. They both contribute towards companies and enterprises owning their whole stack - from the model level all the way up to the context level

💬21🔄11❤️131👀14947📊47

Sierra isn't the first to build this - Ramp, Stripe, CoinBase also have If you want an open source ...

Harrison Chase (@hwchase17) — Fri, 10 Jul 2026 15:04:11 GMT

Sierra isn't the first to build this - Ramp, Stripe, CoinBase also have

If you want an open source version - check out OpenSWE: github.com/langchain-ai/o…

We use it internally (mostly for coding). Model agnostic, fully OSS but integrates seamlessly with LangSmith for o11y

💬22🔄25❤️491👀120903📊212

love this framing of memory as "proactive"

Harrison Chase (@hwchase17) — Fri, 10 Jul 2026 03:32:23 GMT

love this framing of memory as "proactive"

Colin Francis@colifran_

agent memory has always been reactive. OpenWiki makes it proactive. connect to sources, tell it what you care about, and your agent hits the ground running .

what we're building is really exciting! try it out and get involved, it's open source! 👇github.com/langchain-ai/o…7

💬12🔄3❤️51👀20625📊17

big update to openwiki - it now supports two modes: - code brain: create a wiki for a code base - p...

Harrison Chase (@hwchase17) — Thu, 09 Jul 2026 17:55:21 GMT

big update to openwiki - it now supports two modes:

- code brain: create a wiki for a code base
- personal brain: create a wiki for more general purpose tasks (email, web, etc)

webinar at 11am PST (5 minutes!) where we will be talking about wikis! 

events.langchain.com/webinar/llm-wi…

Brace@BraceSproul

x.com/i/article/2075…

💬8🔄21❤️139👀16711📊35

this is going to be a fun one! clay is one of the most ai-forward thinking companies i know of lot...

Harrison Chase (@hwchase17) — Thu, 09 Jul 2026 16:24:07 GMT

this is going to be a fun one!

clay is one of the most ai-forward thinking companies i know of

lots to learn here if you're in NYC

LangChain@LangChain

Join us for a LangChain + Clay meetup with @palashshah, @jeffbarg, Vyshu Khota, and Soroush Khadem. luma.com/jqif2hti

Palash will break down how he built a self-improving agent at LangChain, LangSmith Engine. He'll dive into how the agent turns production traces into durable memory updates that continuously improve an agent's behavior.

Then, the Clay team will give us a peek into how they built out their agentic eval harness, and more.

Interested in attending? RSVP while spaces last.

💬2🔄5❤️29👀7735📊5

langsmith for coding agents

Harrison Chase (@hwchase17) — Thu, 09 Jul 2026 16:00:46 GMT

langsmith for coding agents

LangChain@LangChain

We built a plugin that traces every Claude Code session straight into LangSmith. Three commands, one JSON block, and every message, tool call, and subagent run shows up as an inspectable trace.

Setup takes about two minutes.

How are you currently debugging Claude Code when it goes off track?

Your browser does not support the video tag.

💬6🔄2❤️36👀8088📊10

Setup evals with… terraform??

Harrison Chase (@hwchase17) — Thu, 09 Jul 2026 14:37:33 GMT

Setup evals with… terraform??

Adam Łucek@AdamRLucek

You've heard of Infrastructure as Code- but agent evals can now ride your existing Terraform setup! I've been using the new LangSmith Terraform provider to auto-provision online evals + monitoring alerts for my agents, and think you should too. Here's how 🧵

💬3🔄13❤️115👀20335📊26

Bigger update coming tmrw, but a bunch of small fixes today!

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 19:30:37 GMT

Bigger update coming tmrw, but a bunch of small fixes today!

Brace@BraceSproul

OpenWiki 0.0.3 just launched with tons of bug fixes and improvements!

Some notable changes:
- 'openai-compatible' provider to support any LLM provider that supports the OpenAI API spec! Thank you @bradhuffman 
- deterministically skip updates if no code changes have been made. Shoutout @hudushy 
- GitLab CI example workflow for running update jobs
- and many more smaller fixes and improvements!

lots more coming soon to OpenWiki. also very interested to hear what people want in a memory system

💬3🔄4❤️30👀7213📊7

i want to build a more opinionated/detailed coding experience (for prompt optimization, eval set cre...

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 19:00:58 GMT

i want to build a more opinionated/detailed coding experience (for prompt optimization, eval set creation, some more things)

what is the best way to package it all up?

💬18🔄1❤️16👀5192📊18

Guess what we added someone else to this webinar: @jeffreyhuber, ceo and cofounder of chroma What ...

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 18:18:28 GMT

Guess what we added someone else to this webinar:

@jeffreyhuber, ceo and cofounder of chroma

What does a vector database think about the concept of “wikis”?

Come find out! It’s turning into a party

💬4🔄4❤️41👀7105📊11

Congrats to prime intellect! Love partnering with them on LangChain labs work

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 18:11:29 GMT

Congrats to prime intellect! Love partnering with them on LangChain labs work

Prime Intellect@PrimeIntellect

Announcing our $130M Series A to build the Open Superintelligence Stack

Led by Radical Ventures, with NVIDIA, Intel Capital, Dell Capital, and existing investors

Train, deploy, and continuously improve your own models using our stack.

Own your intelligence.

Your browser does not support the video tag.

💬6🔄5❤️62👀10506📊10

Great opportunity!

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 16:42:14 GMT

Great opportunity!

Gabe Pereyra@gabepereyra

We are hiring for @Harvey’s model training team.

This team will help Harvey expand from the application layer into the model layer and from legal into high end knowledge work more broadly.

We are hiring AI researchers of all seniority, particularly those with experience post-training frontier or open source models.

Our program is centered around large-scale model training, synthetic data generation, long horizon reinforcement learning, and rigorous evaluation in real world deployments. We are scaling-pilled and believe that nothing beats the combination of larger models and better training data.

We’ve been able to generate incredibly realistic legal environments and validated that this allows us to post-train open source models to achieve frontier performance with agents. We plan to scale up these data generation and training efforts significantly across legal to start, and eventually other verticals.

As a researcher, you will have access to thousands of GPUs and unique training data from our product and customer relationships.

Your research will inform Harvey’s product strategy and power AI used for some of the most economically and societally impactful work in the world.

💬0🔄1❤️35👀8132📊5

Love partnering with baseten to make sure everyone can use open weight models in deep agents

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 15:34:47 GMT

Love partnering with baseten to make sure everyone can use open weight models in deep agents

Philip Kiely@philipkiely

x.com/i/article/2074…

💬9🔄6❤️63👀10383📊17

Really excited to partner with @nvidia on the NemoClaw Deep Agents Blueprint Deep Agents is a fully...

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 15:11:59 GMT

Really excited to partner with @nvidia on the NemoClaw Deep Agents Blueprint

Deep Agents is a fully open source agent harness that we are tuning to make perform incredibly well with open models

LangChain@LangChain

Introducing the NemoClaw Deep Agents Blueprint, a reference architecture for building open agent systems developed with @NVIDIA

✅ A fully open stack enterprises can own and customize
✅ Benchmark-leading performance
✅ Over 10x lower inference costs

Blog: langchain.com/blog/langchain…

Video: youtu.be/Yy3JH6dDugc

Your browser does not support the video tag.

💬10🔄18❤️110👀14052📊23

gpt-5.6 sol isnt the only thinking launching thursday we're also releasing a big update to OpenWiki...

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 14:48:33 GMT

gpt-5.6 sol isnt the only thinking launching thursday

we're also releasing a big update to OpenWiki (auto create wikis of code bases... and more?)

we're also going live with a webinar to talk all things wiki: events.langchain.com/webinar/llm-wi…

(openwiki: github.com/langchain-ai/o…)

💬8🔄15❤️91👀13982📊31

agents for ad spend

Harrison Chase (@hwchase17) — Wed, 08 Jul 2026 14:46:54 GMT

agents for ad spend

Michelle Lim@michlimlim

Introducing Flint Ads Agent: we optimize your Google Ads spend for you.

Our agent figures out what’s costing you conversions, using our comprehensive data and optimization engine. Then, it executes the fix for you, and learns from the results.

@BoomPopHQ 10x’ed conversions with it.

Your browser does not support the video tag.

💬3🔄0❤️20👀5308📊5

is anyone standardizing on ATIF as a format for agent traces? or a different format? or is it still ...

Harrison Chase (@hwchase17) — Tue, 07 Jul 2026 23:50:54 GMT

is anyone standardizing on ATIF as a format for agent traces? or a different format? or is it still wild wild west, roll your own

💬15🔄2❤️42👀11643📊24

deepagents is our newest open source project - an open source, model agnostic agent harness this is...

Harrison Chase (@hwchase17) — Tue, 07 Jul 2026 17:35:30 GMT

deepagents is our newest open source project - an open source, model agnostic agent harness

this is maybe the most important academy course we've launched

LangChain@LangChain

🎓 New course launch from LangChain Academy: Introduction to Deep Agents

✅ Learn what a harness is, and why agents need one
✅ Understand the 4 core capabilities of a harness
✅ Start building with Deep Agents
✅ Trace and deploy with LangSmith

Your browser does not support the video tag.

💬8🔄18❤️210👀34795📊51

llm wikis are a glimpse of the future of what agent memory looks like this blog i wrote resonated w...

Harrison Chase (@hwchase17) — Tue, 07 Jul 2026 17:15:00 GMT

llm wikis are a glimpse of the future of what agent memory looks like

this blog i wrote resonated with a lot of folks

will be discussing this (as well as some updates ive made to my beliefs since this!) this thursday with @BraceSproul@devstein64events.langchain.com/webinar/llm-wi…

Harrison Chase@hwchase17

x.com/i/article/2071…

💬22🔄21❤️158👀29656📊56

How a local-first personal AI agent uses orchestration, memory, tools, LangGraph, background workflo...

Harrison Chase (@hwchase17) — Tue, 07 Jul 2026 17:11:07 GMT

How a local-first personal AI agent uses orchestration, memory, tools, LangGraph, background workflows, and child agents

𓁟 SYD 🛸@SydSachar

x.com/i/article/2074…

💬3🔄11❤️63👀13119📊19

Good blog from viv on how improving agents (via rl, harness Eng, anything) boils down to a data mini...

Harrison Chase (@hwchase17) — Tue, 07 Jul 2026 15:05:21 GMT

Good blog from viv on how improving agents (via rl, harness Eng, anything) boils down to a data mining problem over traces

Viv@Vtrivedy10

x.com/i/article/2073…

💬14🔄22❤️160👀35319📊54

devin has a hot take that wikis are NOT the right abstraction should be a fun webinar!!

Harrison Chase (@hwchase17) — Mon, 06 Jul 2026 17:07:05 GMT

devin has a hot take that wikis are NOT the right abstraction

should be a fun webinar!!

Devin Stein@devstein64

does your agent need a wiki? should humans and agent use the same wiki? should you have one wiki or many wikis? 

if you're wiki-curious, join @BraceSproul@hwchase17 and I on Thursday! 

you don't want to miss it  @dosu_ai 🤝 @LangChain

💬8🔄5❤️64👀21111📊15

🚨Emergency webinar: LLM Wikis and how to give your agent memory LLM Wikis are so hot right now - O...

Harrison Chase (@hwchase17) — Mon, 06 Jul 2026 16:14:08 GMT

🚨Emergency webinar: LLM Wikis and how to give your agent memory

LLM Wikis are so hot right now - OpenWiki by @BraceSproul up to nearly 7k GitHub stars in less than a week

I'll be chatting with Brace and @devstein64 about Wikis this Thursday:events.langchain.com/webinar/llm-wi…i

💬24🔄16❤️101👀23937📊46

anyone looked into or using Open Knowledge Format (https://t.co/b2MZgpUx2O) for their wikis?

Harrison Chase (@hwchase17) — Sun, 05 Jul 2026 16:24:09 GMT

anyone looked into or using Open Knowledge Format (github.com/GoogleCloudPla…) for their wikis?

💬57🔄31❤️362👀43199📊173