The pitch: You want Claude running 24/7 in your team's Slack — triaging support questions, investigating bugs, filing GitHub issues — not just when someone has a terminal open. ClaudeClaw makes that a `git clone` + `/setup` away.

What makes it interesting:

🔒 OS-level sandboxing, not Docker. Each agent spawns in <10ms using Anthropic's `sandbox-runtime` (Seatbelt on macOS, bubblewrap on Linux). Kernel-enforced filesystem and network restrictions. The security model inverts Docker's: credentials are passed directly, but the agent can only connect to Anthropic's API — nowhere to exfiltrate.

🧩 Composable extension system. Channels and agents aren't bundled — they're installable extensions with their own repos. `/install-extension slack` clones from GitHub, compiles, copies skills, restarts. No merge conflicts, no monolith. Core stays lean (~8K lines).

🧠 Structured memory. Daily logs, topic files, long-term CLAUDE.md, searchable archive. PreCompact hooks auto-save conversations before context compaction. QMD semantic search (hybrid BM25 + vector + LLM re-ranking) on the roadmap.

📡 Three trigger modes. Channel messages (@mention), scheduled tasks (cron), and webhook triggers (HTTP POST with HMAC-SHA256). CI failures, GitHub events, and monitoring alerts can all invoke agents.

🔧 Self-improving. The entire codebase fits in Claude's context window. Want to change behavior? Tell Claude. It reads, modifies, and commits code changes. No config files — code is the configuration.

Architecture at a glance:

```

Channels → SQLite → Polling loop → Sandbox (Claude Agent SDK) → Response

Webhooks → HTTP → HMAC verify → Queue

```

Single Node.js process. Channels self-register at startup. Per-group isolation: each group gets its own sandbox, memory, filesystem, and agent config (model, tools, cost limits).

Origin: Fork of NanoClaw, rebuilt as a Claude Code plugin in two days. Added sandbox runtime, extension system, message router with hook pipeline, structured memory, webhook triggers, cost tracking, and per-group agent configuration.

Roadmap:

- Multi-model routing (OpenAI, Google, Ollama — provider-prefixed model strings)

- QMD semantic memory (fully local, no API calls for search)

- Obsidian vault integration

- TUI for instance management

Links:

- 🔗 Repo: github.com/sbusso/claudeclaw

- 📄 License: MIT

- 📋 Requirements: macOS/Linux, Node.js 20+, Claude Code, your own API key OAuth

As the leaves change hues and the nights grow crisp, mid-October brings us a blend of nostalgia and innovation, but AI innovation and the LangChain ecosystem continue to thrive. Dive in with us as we unwrap the latest developments in the LangChain universe.

Subscribe now

Projects

LangServe A library that helps developers deploy LangChain runnables and chains as a REST API. LangServe offers features such as automatic schema inference, efficient endpoints, and support for popular Python libraries. It simplifies language processing workflows and provides developers with a seamless way to integrate LangChain functionalities into their projects.

OpenAI Python Library A beta version of the OpenAI Python library version 1.0 is now available, offering new features and breaking changes for users to test before the official release. Read the full release notes and keep up with the discussion.

LobeChat is an open-source chatbot framework that offers a dynamic plugin system, a repository of prompt agents, seamless mobile and desktop experience via PWA, customizable themes, and optimized mobile interactions

Openllmetry built on OpenTelemetry by Traceloop, enhances observability in LLM applications, supports integration with platforms like Datadog and Honeycomb, and offers specialized extensions for tools like OpenAI and Vector DBs; it can be quickly initiated with the Traceloop SDK.

Funcchain offers a pythonic approach to crafting cognitive systems, using pydantic models and langchain to easily integrate llms into apps, optimizing with OpenAI Functions and JSONFormer compatibility.

Videos

James Briggs takes viewers on a journey from novice to expert in LangChain streaming, highlighting the use of GPT-3.5-turbo through ChatOpenAI and its integration with FastAPI.

Sam Witteveen's highly informative and well-presented three-part series on Retrieval Augmented Generation (RAG) offers a comprehensive exploration of Self-querying retrieval, parent document retrieval, and hybrid search.

Integrations

Cloudflare introduces Workers AI, a serverless platform allowing developers to effortlessly execute AI models on a global GPU network, prioritizing accessibility, privacy, and a seamless developer experience. Read the full blog about the new service platform here.

n8n has seamlessly integrated LangChain, enabling users to build LLM apps on its intuitive low-code platform swiftly. This fusion offers tailored AI and advanced chatbot functionalities and harnesses the power of 398 connectors for enhanced automation. Users can easily craft modular applications without coding expertise, optimize LLM model interactions, and embed chat widgets. Explore the new integration here.

That’s all for today!

Thank you for being a part of this continuously evolving journey with LangChain! Your passion and backing truly illuminate our community. If this issue struck a chord with you, please share it with others. Together, we only get stronger!

Share

If you believe something was overlooked or have a project in mind, please feel free to connect with us on our GitHub repository or at awesomelangchain[at]kyrolabs.com. We are eager to hear your thoughts.

Stay incredible, and see you in the next update!

This newsletter is crafted with love by the Kyrolabs team.

PS: We are not affiliated with the products or services mentioned in this update.

As we bid farewell to summer, the energy in the tech sphere is palpable! Projects are springing back to life, and the air is thick with innovation. Grab your favorite autumn drink, and let's catch up on all the thrilling updates that came our way.

Subscribe now

Projects

Rivet is an IDE for creating complex AI agents and prompt chaining, and embedding it in your application.

AgentVerse is a flexible framework that simplifies building custom multi-agent environments for large language models (LLMs).

Langstream build and run event-driven Gen AI applications

GitWit is a container-based agent specialized in making useful commits to git repositories.

RestGPT is an LLM-based autonomous agent that can control applications via RESTful APIs

Magentic seamlessly integrate LLMs as Python functions

Promptfoo test your prompts. Evaluate and compare LLM outputs, catch regressions, and improve prompt quality.

Videos

Superagent is a platform dedicated to the development and management of LLM agents, and this webinar delves into the technology that powers it.

James Briggs demonstrates building a chatbot using Retrieval Augmented Generation (RAG) with OpenAI's gpt-3.5-turbo

Hosted by Kelvin Lawerence, this session simplified the use of graphs with Amazon Neptune through Langchain and LLMs.

Articles

The new OpenAI GPT-4V model, short for GPT-4 with vision, allows users to instruct GPT-4 to analyze image inputs, marking a significant stride in AI research by incorporating additional modalities like images into large language models (LLMs). Read the full system card here.

Molly Cantillon for Langchain, demonstrates building a chatbot named Chat LangChain, which provides information about LangChain by indexing and navigating through the Python documentation and API reference, while also delving into its architectural framework. Link to the tutorial here.

Cloudflare published a tutorial on utilizing Langchain and Pinecone to index a Notion workspace, and on developing a custom ChatGPT plugin with Cloudflare Workers for querying purposes. You can read the tutorial here.

AutoGen from Microsoft streamlines the orchestration of large language model (LLM)-based workflows. Through customizable agents, it fosters a collaborative space among LLMs, tools, and humans to efficiently address complex tasks. This development heralds a new era in AI, with more robust and automated LLM applications. Explore the full insight into AutoGen's potential here.

Delve into Fine-tuning GPT-3.5-Turbo to efficiently convert natural language queries into SQL. Read the article here.

OpenAI released a guide for teachers about ChatGPT, including suggested prompts. Read it here.

This article analyzes how users are using ChatGPT including prompt types and commonly used words. Read it here.

The article explores improving Large Language Models (LLMs) performance through dataset-centric fine-tuning. Read it here.

That’s all for this month!

Thank you for being a part of this continuously evolving journey with LangChain! Your passion and backing truly illuminate our community. If this issue struck a chord with you, please share it with others. Together, we only get stronger!

Share

If you believe something was overlooked or have a project in mind, please feel free to connect with us on our GitHub repository or at awesomelangchain[at]kyrolabs.com. We are eager to hear your thoughts.

Stay incredible, and see you in the next update!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we are looking for individuals who are enthusiastic about AI Agents and LangChain. If you're keen on embarking on this exhilarating adventure with us, send us a note at contact[at]kyrolabs.com, sharing your interest in AI Agents and background.

Please note that we are not affiliated with the products or services mentioned in this update.

Subscribe now

Projects

Two big topics in August: Llama2 and fine-tuning.

Meta also introduced Code Llama, a state-of-the-art Large Language Model built on top of Llama 2, available in three specialized models for generating and understanding code and natural language

• AutoChain: A lightweight framework for quickly building and evaluating conversational agents using LLMs, facilitating user experiments and customization.

• Codeinterpreter-api: A LangChain implementation of the ChatGPT Code Interpreter.

• LangSmith-Cookbook: provides tutorials for the LangSmith platform.

• Lagent: A lightweight framework for building LLM-based agents.

• Marvin: Another lightweight AI engineering framework for building natural language interfaces that are reliable, scalable, and easy to trust.

Monitoring

Mona is an AI monitoring platform that provides valuable insights into ML models and intelligent automation by detecting and resolving issues rapidly while reducing alert fatigue by notifying only specific root causes.

Graphsignal is an AI-native observability platform that offers comprehensive monitoring, debugging, and collaborative tools for AI agents and LLM-powered applications

Langsmith from LangChain, a developer platform for LLM applications to test, monitor, and debug your applications

Videos

Discover the capabilities of Llama2 as a conversational agent in LangChain.

Learn how to bring your personality and tone to LLM apps in the Fine-tuning in Your Voice Webinar.

Articles

LangChain's observability makes it easy to fine-tune LLMs in your unique style by converting messaging platform data here

Enhance your performance and profits using OpenAI's 3.5 Turbo fine-tuning with LangSmith here

Fine-tuning GPT-3.5-Turbo for Natural Language to SQL here

That’s all for this week!

Thank you for being a part of this continuously evolving journey with LangChain! Your passion and backing truly illuminate our community. If this issue struck a chord with you, please share it with others. Together, we only get stronger!

Share

If you believe something was overlooked or have a project in mind, please feel free to connect with us on our GitHub repository or at awesomelangchain[at]kyrolabs.com. We are eager to hear your thoughts.

Stay incredible, and see you in the next update!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we are looking for individuals who are enthusiastic about AI Agents and LangChain. If you're keen on embarking on this exhilarating adventure with us, send us a note at contact[at]kyrolabs.com, sharing your interest in AI Agents and background.

Please note that we are not affiliated with the products or services mentioned in this update.

We bring you various updates, projects, and insightful content this week. So let's dive in!

Subscribe now

Updates

Partnership with Streamlit: Exciting news! Langchain has partnered with Streamlit. You can read more about this here: LangChain-Streamlit partnership

Automatic Metadata Extraction: Now, you can leverage an LLM to extract metadata. Check out the details here

Property Extraction with Doctran: Similar to the metadata tagger, the doctran property extractor lets you define the properties to extract from a document's text and add them to its metadata. Get more information here.

KoboldAI Model Wrapper: KoboldAI offers a variety of models fine-tuned for story writing. Read more here

Changelogs:

Langchain Python Changelog

Langchain JS/TS Changelog

Projects

a16z Companion App: Check out the project here

LangChain Streamlit Agent: Discover this new Streamlit agent for the new partnership here

Airtable QNA: Now, ask questions to your Airtable base in natural language. Learn more here.

Videos

LangChain "Characters" Webinar

Image to Speech Pipeline with LangChain and HuggingFace Models: Although a month old, this is an excellent showcase. Watch the video

Claude-2 Meets LangChain!

Articles

Building a Conversational Agent with Memory with Motorhead: Learn how to stack chains as tools here.

Analyzing User Interactions with LLMs to Improve our Documentation: Understand how LLMs can enhance documentation here.

The Emergence of LangChain: Replit explores the rise of LangChain in this article.

That’s all for this week!

Thank you for being a part of this incredible LangChain journey with us. Our community's vibrancy is a credit to you all. If you enjoyed this week's digest, feel free to share it. The more, the merrier!

Share

If you think we missed something or if you have a project to suggest, don't hesitate to reach out on our GitHub repository or email us at awesomelangchain[at]kyrolabs.com. We look forward to hearing from you.

Keep being awesome, and see you next week!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we seek individuals passionate about AI Agents and LangChain. If you're interested in joining us on this exciting journey, drop us a message at contact[at]kyrolabs.com and tell us about your interest in AI Agents and your background.

Welcome to Week 28 of Awesome LangChain updates! Even though many of us are kicking off our summer holidays, plenty's still happening in the LangChain world. Let's dive straight in and check out what's new. Enjoy!

Subscribe now

Integrations

New integrations are now available directly on the dedicated website with some new integrations with Office365 and the Changelogs:

Langchain Python Changelog

Langchain JS/TS Changelog

Awesome Projects

• BlockAGI: BlockAGI conducts iterative, domain-specific research and outputs detailed narrative reports to showcase its findings

• A fork of the Vercel Chatbot starter kit, this time with Supabase infrastructure:

• AilingBot: Quickly integrate applications built on Langchain into IM, such as Slack, WeChat Work, Feishu, DingTalk.

Articles

• Building AI data pipeline with Langchain and Airbytes

• Great article/code on creating an HR agent.:

  Code: Autonomous HR Chatbot
  Article: Creating a (mostly) Autonomous HR Assistant with ChatGPT and LangChain’s Agents and Tools

• Another combo with Running Large Language Models on Google Cloud Platform:

  Code - Article

  

Videos

• This week DeepLearning.AI released a new video presented by Harrison Chat with your document

• great video from Sam: Using a conversational form with LLMchain and functions

Other

Enjoy a LangChain cocktail made by BarGPT

Wrapping Up!

Thank you for being a part of this incredible LangChain journey with us. Our community's vibrancy is a credit to you all. If you enjoyed this week's digest, feel free to share it. The more, the merrier!

Share

If you think we missed something or if you have a project to suggest, don't hesitate to reach out on our GitHub repository or email us at awesomelangchain[at]kyrolabs.com. We look forward to hearing from you.

Keep being awesome, and see you next week!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we always seek individuals passionate about AI Agents and LangChain. If you're interested in joining us on this exciting journey, drop us a message at contact[at]kyrolabs.com and tell us about your interest in AI Agents and your background.

Let's cool down with some fresh LangChain updates! As the summer rolls on, we hope you're finding time to relax and recharge, even as you keep pace with the dynamic world of LangChain. Whether you're dipping into this week's digest from a sunny beach or a bustling city, we're thrilled to bring you the latest updates, projects, and insights from the LangChain ecosystem.

So, grab a refreshing drink, and let's dive in!

Subscribe now

Integrations

First, a brand new Integration page for LangChain: https://integrations.langchain.com/

• AWS Step Functions Agent Toolkit

• Prem Prem provides a unified environment to develop AI applications and deploy AI models on your infrastructure

• New text splitters, including a Context-aware version

Changelog

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

And if you are looking for support for other languages, some active development exists in ruby, golang and java.

Awesome Projects

• FunctionChain: simplifies and organizes the process of invoking OpenAI functions in your Node.js applications

• Marvin: A batteries-included library for building AI-powered software

• GPT Migrate: Easily migrate your codebase from one framework or language to another.

Articles

Articles worth a read:

• A Hands-on Journey to a working LangChain LLM Application

• What is Langchain and why should I care as a developer?

Videos

MPT-30B Chatbot with LangChain! by James Briggs

Walkthrough setting up a MPT-30B on a single GPU (A100, Pro account required) and chat with Langchain and Huggingface

Webinar on LangChain: Chains vs Agents

We've had some fantastic video content this week, particularly packed with valuable insights about Agents and design examples. Not to be missed is @swyx's compelling presentation on 'smol developer'. Dive in to expand your knowledge!

So, what is your choice, Chain or Agent?

Tweets

Making LangChain more interoperable with Open Source models backends

https://twitter.com/hwchase17/status/1673393425456963611

Wrapping Up!

Thank you for being a part of this incredible LangChain journey with us. Our community's vibrancy is a credit to you all. If you enjoyed this week's digest, feel free to share it. The more, the merrier!

Share

Your feedback is vital to us. If you think we missed something or if you have a project to suggest, don't hesitate to reach out on our GitHub repository or email us at awesomelangchain[at]kyrolabs.com. We look forward to hearing from you.

Keep being awesome, and see you next week!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we always seek individuals passionate about AI Agents and LangChain. If you're interested in joining us on this exciting journey, drop us a message at contact[at]kyrolabs.com and tell us about your interest in AI Agents and your background.

We've seen a flurry of activity this week, primarily focusing on OpenAI's new features, which we mentioned last week. Functions pique interest and branch out in numerous directions, like extractions and faster agents.

Subscribe now

Here are some updates from the Langchain ecosystem you should know about:

Integration

Langchain Extraction It uses OpenAI's functions to extract structured information from content.

Mongo Vector database: Mongo announced their vector search offering LangChain already got integrations with it.

Motherduck is a managed DuckDB-in-the-cloud service.

VertexAI: We have further integrated more models with VertexAI.

Changelog

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

Awesome Projects

Following last week's release of AI from Vercel, this week, they shared ChatBot :

• Vercel Labs' AI Chatbot: A feature-rich, hackable Next.js AI chatbot.

• a16z's AI Getting Started: A JavaScript AI starter stack for weekend projects, complete with image/text models, vector stores, auth, and deployment configs.

• Embedchain: A framework for creating LLM-powered bots over any dataset.

Articles & Videos

Articles & Videos Worth Your Time:

• "Build a Question-Answering over Docs SMS Bot with LangChain in Python" featured on the Twilio blog. A special shout-out to the Twilio team for offering a free trial for everyone to give it a whirl. You can utilize the promo code AWESOMELANGCHAIN23 for access (find instructions here).

• Gil Fernandes' series on Medium, particularly the article "LangChain's Router and Callbacks."

• The extensive one-page guide about Langchain on Shurutech.

• “LlamaIndex: The Ultimate LLM Framework for Indexing and Retrieval" on Towards Data Science.

Videos

• LangChain "OpenSource LLMs" Webinar, from previous week

• And this week, another podcast with great guests about functions:

• Sam Witteveen video: Tagging and Extraction - Classification using OpenAI Functions

• Using ChatGPT with your own Data

From Twitter

• LLMops on Google Cloud micro-services with

  https://twitter.com/holomarked/status/1672669761669300225?s=12&t=79HZhE3ChdH7091ieZLL1A

• Langchain Ecosystem by Lance Martin

  https://twitter.com/RLanceMartin/status/16722906706381660

Other News

Check out a16z's article on "Emerging Architectures for LLM Applications."

That's All for Now!

That's it for this week!

As always, we want to thank you for being a part of this exhilarating journey with us. The vibrancy of the LangChain community is a testament to each of you. If you found this week's digest helpful, we encourage you to share it. More voices lead to richer and more varied content.

Share Awesome Langchain

We cherish your feedback. If you believe we missed something or want to suggest a project, feel free to create a PR or an issue on our GitHub repository or drop us a line at awesomelangchain[at]kyrolabs.com.

Stay curious, stay awesome, and see you next week!

This newsletter is crafted with love by the Kyrolabs team.

PS: At Kyrolabs, we always seek individuals passionate about AI Agents and LangChain. If you're interested in joining us on this exciting journey, drop us a message at contact[at]kyrolabs.com and tell us about your interest in AI Agents and your background.

This week, we had a big OpenAI update bringing functions and 16k models. As usual, Langchain was quickly updated, and we already have some interesting content to start using them.

Subscribe now

Quick highlight on some Langchain updates:

Document loaders

• Airtable

• Unstructured XML

Integration

On the same day OpenAI released new models and features, Langchain introduced OpenAI Functions in LangChain. Besides empowering agents, it also allows extracting structured information from unstructured docs.

Changelog

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

Awesome Projects

Checkout those awesome projects showcasing the versatility of LangChain:

Vercel AI SDK: Build AI-powered applications with React and Svelte

Robby Chatbot: AI chatbot 🤖 for chatting with CSV, PDF, TXT files 📄 and YTB videos🎥, powered by Langchain, OpenAI, PyTube, and Streamlit.

DocNavigator: AI-powered chatbot builder that is designed to improve the user experience on product documentation/support websites

Multi-Modal LangChain agents in Production: Deploy LangChain Agents and connect them to Telegram

Alternatives

OpenLLM: An open platform for operating large language models (LLMs) in production. Fine-tune, serve, deploy, and monitor any LLMs with ease.

FlagAI: FlagAI (Fast LArge-scale General AI models) is a fast, easy-to-use, and extensible toolkit for large-scale models.

Articles & Videos

Sam Witteveen Videos

• What can you do with 16K tokens in LangChain?

• OpenAI Functions + LangChain: Building a Multi-Tool Agent

Greg Kamradt LangChain cookbook part 2 Use cases:

• From Sequoia article giving an overview of LLMs stack usage in their portfolio:

  “Nearly every company in the Sequoia network is building language models into their products.“
  “Every practitioner we spoke with said AI is moving too quickly to have high confidence in the end-state stack, but there was consensus that LLM APIs will remain a key pillar, followed in popularity by retrieval mechanisms and development frameworks like LangChain.“

  Note: Sequoia led the LangChain funding round.

From Twitter

• Deploy your bot platform using Dialoqbase, Railway, and Supabase.

Other News

Check out the week program with over $850k in credits from Vercel and top AI platforms like LangChain. https://vercel.com/blog/vercel-ai-accelerator

That's All for Now!

Thank you for being part of this exciting journey with us. The LangChain ecosystem wouldn't be the same without its vibrant community. If you enjoyed this week's digest, please consider sharing it with others. With more people involved, our content becomes richer and more diverse.

Share Awesome Langchain

Your input is invaluable to us. If we missed something or you'd like to suggest a project, please make a PR, add an issue on our GitHub repository, or email us at awesomelangchain[at]kyrolabs.com. We can't wait to hear from you!

Stay curious, stay awesome, and see you next week!

The team at Kyrolabs lovingly curates this newsletter.

PS: Looking for an opportunity to work with AI Agents and LangChain? At Kyrolabs we are looking for passionate people to join the journey, if you are interested, drop us a message at contact[at]kyrolabs.com and share with us your interest in AI Agents and background.

This week, we're delving into personal use cases of LangChain with insightful articles and intriguing videos about Quivr, AnythingLLM, and OpenDAN.

Langchain Updates

Quick highlight on some Langchain updates:

• Snowflake: Load data from your Snowflake into document objects. This will allow you to split, embed, and eventually query it with a semantic search.

• Unstructured CSV Loader:

• Lord of the Retrievers: Merger Retriever allows for an easy combination of MULTIPLE retrievers.

• Update for GraphCypherQAChain: Limit the number of results, Return intermediate, and Return Direct results.

Changelog

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

Awesome Projects

Checkout those awesome projects showcasing the versatility of LangChain:

• GoT Chatbot: Repo for creating GoT Chatbots (ex: talk with Tyrion Lannister)

• Anything LLM: A full-stack application that turns any document into an intelligent chatbot with a sleek UI and an easier way to manage your workspaces.

• Salute: A simple and declarative way to control LLMs

• OpenDAN: open source Personal AI OS consolidates various AI modules in one place for your personal use.

• ReLLM: Regular Expressions for Language Model Completions.

• Dialoqbase: a web application that allows you to create custom chatbots with your knowledge base

• CSV AI: Ultimate app powered by LangChain that allows you to unlock hidden insights in your CSV files

Articles & Videos

• Multilingual Chat Bot for Personal Documents (medium article)

• Langchain Video, Quivr Deepdive, Your second brain with Langchain

3 episodes by Sam Witteveen

• Output Parser

• Custom Medical Agent with Memory

• Understanding ReACT

From Twitter

• Overview of VectorDBs

• A thread about Managing LLM limitations

Other News

Over the weekend, prominent scientists Andrew Ng and Geoffrey Hinton reached out to the scientific community, urging for a unified agreement on potential risks similar to the stance maintained by climate scientists. This crucial discussion, happening amidst a climate of differing opinions, has also drawn the interest of Yann LeCun:

• Andrew Ng's tweet

Regarding our favorite framework, Andrew Ng wrote: "As for making LLMs plan, OpenAI's process supervision, ReAct-style prompting, and 'wild west' stuff like LangChain agents feels like tons of people are experimenting with different ideas."

That's All for Now!

Thank you for being part of this exciting journey with us. The LangChain ecosystem wouldn't be the same without its vibrant community. If you enjoyed this week's digest, please consider sharing it with others. With more people involved, our content becomes richer and more diverse.

Share Awesome Langchain

Your input is invaluable to us. If we missed something or you'd like to suggest a project, please make a PR, add an issue on our GitHub repository, or email us at awesomelangchain[at]kyrolabs.com. We can't wait to hear from you!

Stay curious, stay awesome, and see you next week!

The team at Kyrolabs lovingly curates this newsletter.

Subscribe now

Langchain Updates

Document loaders

• NotionDB ported to JS/TS

• Spark DataFrame

• Trello

• Github issues and PRs

Integrations

• Momento Cache, to reduce costs

• Twilio to send SMS, make sure that your agent does not start chatting with its peers! And an article demoing Build a Basketball SMS Chatbot with LangChain Prompt Templates in Python.

• C Transformer Interface opens the door to connect LLMs ported by Georgi Gerganov running on lower-cost machines (like llama.cpp).

• Qdrant improvements: Richer filtering support, Self-querying retrieval, Batched upserts

• Brave private search API integration

• Amazon Bedrock Models: now has LLM and embedding model wrappers for Amazon's new Bedrock service, which serves both Amazon's own Titan models and 3rd party models like Anthropic's Claude.

• Arguilla open-source data curation platform for LLMs, empowering data teams from data labeling to model monitoring.

Output Parsers

• Enum and Datetime

Agents

Multi-dataframe agents shows how the agent can interact with multiple csv files passed in as a list.

Generative Agents This script implements a generative agent based on the paper Generative Agents: Interactive Simulacra of Human Behavior by Park, et. al.

Changelog

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

Awesome Projects

Checkout those awesome projects showcasing the versatility of LangChain:

Gorilla: An API store for LLMs. Enables LLMs to use tools by invoking APIs. Given a natural language query, Gorilla comes up with the semantically- and syntactically- correct API to invoke.

Voyager: An LLM-powered embodied lifelong learning agent in Minecraft that continuously explores the world, acquires diverse skills and makes novel discoveries without human intervention.

WingmainAI: A tool for interacting with real-time transcription of both system and microphone audio.

Local GPT: Inspired by Private GPT with the GPT4ALL model replaced with the Vicuna-7B model and using the InstructorEmbeddings instead of LlamaEmbeddings

Articles & Videos

This week's picks include articles on implementing GPT4All with LangChain, training your own LLM using privateGPT, and LangChain Autonomous Agent - BabyAGI. Don't miss the collection of videos, including a comprehensive course on LangChain and an overview of the state of GPT by Andrej Karpathy.

• Implementing GPT4All Locally with Python and Langchain

• Training Your Own LLM using privateGPT: Learn how to train your own language model without exposing your private data to the provider

• LangChain Autonomous Agent — BabyAGI , integrated as Web App

• Auto-Evaluation of Metadata Filtering by Lance Martin looks at how to improve retrieval using metadata filtering.

• Interesting thread from Harisson on condensing chat history to improve performances and reduce cost using 2 models: https://twitter.com/hwchase17/status/1663576081918742530

• And a visual overview of LangChain retrievers:

source: https://twitter.com/pwang_szn/status/166312305009794662

New LangChain course

When Harrisson collaborates with Andrew Ng, we get another awesome piece of content with this 1-hour course on LangChain. Check it out:

🎥 LangChain for LLM Application Development

State of GPT

One of the more comprehensive videos of this year about LLMs and prompt engineering. Summarize so many notions and concepts developed recently and a good overview and understanding of building LLMs. Andrej Karpathy is, as usual, very good at sharing his knowledge.

🎥 link to the video

Converting a LangChain App from OpenAI to OpenSource

Another great video from Sam Witteven walks us through building a retrieval QA application with Chroma and OpenAI, then converting it to open source models.

🎥 link to the video

Testing LLM with GPT4All

🎥 Testing Free Large Language Models with Steaminlit, LangChain and GPT4All by Nicholas Renotte

Other News

Is LangChain the new LAMP stack? THENEWSTACK takes on the incredible growth of langChain: LangChain: The Trendiest Web Framework of 2023, Thanks to AI

The open source Vector Toolkit for Postgres, Use the Supabase client libraries to store, index, and query your vector embeddings at scale. https://supabase.com/vector

That's All for Now!

Thank you for being a part of this exciting journey with us. The LangChain ecosystem wouldn't be the same without its vibrant community. If you enjoyed this week's digest, please consider sharing it with others. With more people involved, our content becomes richer and more diverse.

Share Awesome Langchain

Your input is invaluable to us. If we missed something or you'd like to suggest a project, please make a PR, add an issue on our GitHub repository, or email us at awesomelangchain[at]kyrolabs.com. We can't wait to hear from you!

Stay curious, stay awesome, and see you next week!

The team at Kyrolabs lovingly curates this newsletter.

We've closely followed Langchain for months while working on our AI LLM project. We've found ourselves resonating with its vision and progress, so we've decided to keep track of Langchain's ecosystem and share it with all of you.

In this newsletter, we're bringing you updates about Langchain, spotlighting new projects, and sharing other content that showcases how Langchain is being used. We aim to help you stay in the loop with all the exciting developments.

We're planning to send these updates to your inbox every week. Your feedback matters greatly to us, so feel free to share your thoughts.

Subscribe now

Here's to an exciting journey through Langchain!

Langchain Updates

Dive into the latest developments in Langchain:

New Integrations

• Neo4j Integration and GraphCypherQAChain

• OpenLM API: a lightweight package that lets you interface with multiple model providers using the OpenAI API (no streaming yet)

• Typesense Vector store: Typesense is an open-source, in-memory search engine supporting full-text and semantic search

• Add support for Open Source models through MosaicML API

• Monitoring with Whylabs

• Psychic as a data loader (an open-source universal document loader)

New Loaders

• Vertex AI Support: Google AI service

• Iugu Document Loader: Iugu is a Brazilian service and software as a service (SaaS) company

• Joplin Document Loader: open-source note-taking app

• Bibtex Document Loader: file format and reference management system

• Mastodon loader

Toolkits

And a list of new powerful toolkits:

Create multi-modal agents powered by Azure Cognitive Services

Weaviate Self-Query Retriever: let an LLM turn questions into structured queries against a DB

Use GPT4all again: how to use LangChain to interact with GPT4All models.

Chat Agent for Structured Tools: work with tools that take an input conforming to an arbitrary object schema

This is a curated list of updates. For a comprehensive look at all updates on both Python and JS, review here:

Langchain Python Changelog

Langchain JS/TS Changelog

Articles

Deepen your Langchain knowledge with these insightful articles:

• Privacy-first AI search using LangChain and Elasticsearch

• Deploy a Voice-Based Chatbot with BentoML, LangChain, and Gradio

• Create a Code Interpreter Chatbot with Pyodide, LangChain, and OpenAI

• LangChain has added Cypher Search

• LangChain tutorial at PromptChap

Videos

• Langchain Retrieval Webinar

• Sam Witteveen videos (Highly recommended channel)

  • Information Extraction with LangChain & Kor

  • Camel + LangChain for Synthetic Data & Market Research

• Using NEW MPT-7B in Hugging Face and LangChain by James Briggs

• The easiest way to deploy privateGPT app

Awesome Projects

These are the awesome projects added this week to the list:

• Chainlit: Build Python LLM apps in minutes ⚡️

• Airtable-QnA a question-answering tool for your Airtable content

• ix Autonomous GPT-4 agent platform

• Psychic: Universal APIs for unstructured data.

• Zep: Zep: A long-term memory store for LLM / Chatbot applications

• DB-GPT: Interact your data and environment using the local GPT

• SuperAgent: Deploy LLM Agents to production

• snowChat ❄️: Chat with you're snowflake

• LangchainJS Worker: LangchainJS worker on cloudflare

• DuetGPT: A conversational semi-autonomous developer assistant.

• Graphsignal Plugin: Observability for AI agents and LLM-powered applications. Trace, monitor, and debug LangChain in production.

Alternatives

• e2b: Open-source platform for building & deploying virtual developers’ agents

Other news

This week, Qdrant has added native support for authentication in v1.2.0. This is a fantastic addition if you aim to use this vector database in production without managing another layer of security.

That's All for Now!

Thank you for being part of this incredible community! Each of you helps improve Awesome Langchain, and we appreciate your contributions. If you found this newsletter helpful, please consider sharing and forwarding it. The more people we engage, the richer and more diverse our content will be.

Share Awesome Langchain

Missed some content or want to suggest a project? Don't hesitate to make a PR, add an issue on our GitHub repository, or drop us a line at awesomelangchain[at]kyrolabs.com. We'd love to hear from you! Stay curious, stay awesome!

The team at Kyrolabs lovingly curates this newsletter.

Subscribe now

Subscribe now