Misbah Syed @MisbahSy Twitter profile | Pikagi

Pikagi

Misbah Syed

@MisbahSy

4,713

Followers

745

Following

263

Media

1,206

Statuses

AI Builder & Educator. DevRel @langflow_ai . Builder @MenloParkLabsAI

Vancouver, Canada

https://t.co/GYLz0iiPIZ

Joined August 2015

Don't wanna be here? Send us removal request.

Pinned Tweet

@MisbahSy

Misbah Syed

3 months

🚀 Introducing Brainy Documents! 🧠📄 Generate explainer videos for your PDFs using AI. Drop a PDF and let AI generate videos explaining the content for you. Built using @LangChainAI @DeepgramAI @UnstructuredIO BONUS: Brainy Documents can also generate presentation slides.

8

41

228

Last Seen Profiles

@bgarciate

@IPG_Kris

@takemytalent

@Cs_Pinto

@illli18_

@papmantan69_

@maskartists

@visitagnes1770

@_Pista2021_

@MadameLumanii

@jww3l

@AhmedE73254

@Dxiansheng00

@ChimaSo39338602

@andrew_sorrells

@Evershite

@Diyarbakirmem

@bokeplokalmalam

@connectfarm1

@_Mirzapuri_

@setataula

@JungleTopBrand

@MIKINGyt

@Katerationopia

@geongi4

@manshinryosho

@Abd_LPB

@sserhatkeskin

@V5b5gB

@javan_bowen

@Shaitannegro

@yoonuas

@Daniel_clifton7

@memakker2

@Tinajam14043176

@TobiasSeck

@MisbahSy

Misbah Syed

1 year

Did you know @CohereAI open-sourced 94 million vectors of Wikipedia embeddings from 10 languages? And @weaviate_io hosted 10 million of them to search for free? With a few lines of code using @LangChainAI you can retrieve them. But we face one issue, let's solve 🧵↓

Tweet media one

22

163

974

@MisbahSy

Misbah Syed

1 year

With this modified code, you can now upload documents in multiple formats (.pdf, .txt, .docx, .pptx, .jpg, .png, .eml, .html) and semantically search in 100+ languages supported by @CohereAI . Embeddings saved @qdrant_engine . Flask-backend uses @LangChainAI . Code open-sourced.🧵👇

Tweet media one

29

115

808

@MisbahSy

Misbah Syed

1 year

Let's build a @LangChainAI agent in under 30 minutes that can access the internet, chat with a PDF, and help us understand code. All without a single line of code using @FlowiseAI and @bubble . This agent will have access to 3 tools and a memory as shown. Tools explained in 🧵👇

Tweet media one

30

151

798

@MisbahSy

Misbah Syed

5 months

An attempt to solve the LLM challenge by @karpathy Built an LLM-based app using @LangChainAI to go from a YouTube link to a nicely written markdown article. (still a work in progress)

@karpathy

Andrej Karpathy

5 months

Fun LLM challenge that I'm thinking about: take my 2h13m tokenizer video and translate the video into the format of a book chapter (or a blog post) on tokenization. Something like: 1. Whisper the video 2. Chop up into segments of aligned images and text 3. Prompt engineer an LLM

213

371

5K

18

69

709

@MisbahSy

Misbah Syed

1 year

Deploy a PrivateGPT chat-with-documents app in less than 15 mins. PrivateGPT's focus on no-data leaks makes it ideal for enterprises and organizations. Built on @LangChainAI , this app utilizes @huggingface embeddings and GPT4ALL for RetrievalQA. 🧵↴

Tweet media one

12

102

614

@MisbahSy

Misbah Syed

5 months

🚀 Introducing RAGTune: An Open-Source tool for tuning and optimizing RAG pipelines! Curious about the best combo of LLMS, embeddings, retrievers, etc. for your RAG app? Now you can easily experiment and see what works best. 1/ Start with uploading documents and providing some

32

81

521

@MisbahSy

Misbah Syed

5 months

Prob: There are too many AI papers to read (and a bit difficult to understand) Sol: A PDF2Video explainer. Drop a PDF and get a well-explained video incl. text, images, equations, figures, etc. Uses @LangChainAI , @UnstructuredIO , @DeepgramAI 1/ Attention is all you need:🔈🔊

14

104

549

@MisbahSy

Misbah Syed

1 year

How about a webapp for privateGPT? Now you can deploy privateGPT as a webapp using fastAPI backend & @streamlit frontend with the code below. Repo link in thread. Uses @huggingface embeddings and supports 12 doc formats ingestion using @LangChainAI (see tweet by @ivanmartit )

Tweet media one

@ivanmartit

Iván Martínez

1 year

privateGPT got two major updates: - ingestion time improved 500x 🤯 - 12 document formats supported (pdf, docx, epub, md...) 📄 Learn how it works and test it out at And become part of the privateGPT community!

33

208

1K

13

81

439

@MisbahSy

Misbah Syed

1 year

The future of web design with AI! 🤯 So made a @mckaywrigley inspired video, where I asked Siri to convert a simple figma template into a nice-looking website. All of the work is done by a @LangChainAI agent; code import, modification, and deployment to @vercel 🧵

25

69

382

@MisbahSy

Misbah Syed

11 months

Learning @LangChainAI is one of the easiest and best ways to start building LLM apps. Happy to collab with @hwchase17 and open-source LangChain Teacher, a chat-based tutor app that walks you through the core components of LangChain and their implementation. Link ↓🧵

13

62

340

@MisbahSy

Misbah Syed

1 year

With these few lines of code, you can deploy a backend for multilingual semantic search over a PDF using @CohereAI and @qdrant_engine , with the ability to embed and perform a search in 100+ languages. Certainly, you can see the power of @LangChainAI here.

Tweet media one

7

55

333

@MisbahSy

Misbah Syed

1 year

🔥AI Agents are currently a hot topic in the AI community. With @LangChainAI , you can work with different types of agents out of the box. 1) ReAct Agent: This agent utilizes the Reason-Action framework to determine the appropriate tool based on its description.

6

70

314

@MisbahSy

Misbah Syed

1 year

Prompt Templates can be quite powerful if used the right way. Let's explore three types of prompt templates you can use in @LangChainAI . 1) Basic prompt template - Uses a single input from the user and combines it with the template string to send the combined prompt to the LLM.

14

58

283

@MisbahSy

Misbah Syed

1 year

Adding Memory is important to help Chains and Agents remember previous interactions. Below are a few memory types you can use for your @LangChainAI apps. 1) ConversationBufferMemory: Keeps a list of the interactions and can extract the messages in a variable.

8

54

269

@MisbahSy

Misbah Syed

1 year

For all no-code/low-code enthusiasts, you can create your own @LangChainAI -based app with ease using @FlowiseAI and @bubble . I made a step-by-step tutorial to walk you through building a PDF Semantic Search app using @OpenAI and @pinecone . 🧵👇

Tweet media one

11

34

249

@MisbahSy

Misbah Syed

1 year

Chains are the core building blocks of LangChain. Below are some of the most important ones you should know if you're building with @LangChainAI . 1) LLMChain: A simple yet powerful chain that consists of a prompt template and a call to an LLM (or a chat model).

3

47

246

@MisbahSy

Misbah Syed

1 year

With these few lines of code, you can build a chat app with Falcon LLM using @LangChainAI and @chainlit_io Video implementation and repo link below ↓

Tweet media one

9

38

246

@MisbahSy

Misbah Syed

1 year

For anyone looking to build a no-code app with @LangChainAI , LangFlow (by @logspace_ai ) offers a drag-and-drop interface and has undergone a complete revamp. Now, you can easily build your own chat-PDF app by following the steps in this video. Link 🧵↓

Tweet media one

6

56

241

@MisbahSy

Misbah Syed

1 year

This is cool! I ran the extraction chain within @LangChainAI on the State of the Union text, works wonders! Can only imagine all the use cases for this.

Tweet media one

@hwchase17

Harrison Chase

1 year

The new @OpenAI functions are good for other things besides agents Another killer use case is extracting structured information from unstructured docs We've adding support for extraction AND tagging in @LangChainAI - thanks to @fpingham for code and @jxnlco for review 🧵

27

113

740

7

37

234

@MisbahSy

Misbah Syed

10 months

For no-code builders looking to make AI apps using @LangChainAI together with hundreds of integrations, @n8n_io 's latest update includes LangChain components built-in. Let's look at some cool AI automation you can do: 1/ Automated newsletter generation from a podcast episode

Tweet media one

10

30

203

@MisbahSy

Misbah Syed

1 year

Launched the test version of the Ask Quran app built using @CohereAI multilingual model, @LangChainAI @pinecone , and @OpenAI GPT-3. The app is now deployed to @vercel (inspired by @steventey ). try at:

21

31

187

@MisbahSy

Misbah Syed

8 months

Use AI to generate social media content for a month. The workflow below can be used to automate content generation & posting. Built w/ @LangChainAI & @n8n_io 1/Drop images (of products, deals, promos) in a drive folder. Call product photos API like @pebblely_ai for variations

Tweet media one

6

29

180

@MisbahSy

Misbah Syed

6 months

Understand LangGraph by @LangChainAI , by building simple graph apps. 🧵↓ Let's start with a simple app with two nodes attached by an edge

Tweet media one

4

29

177

@MisbahSy

Misbah Syed

1 year

"Hey LangChain!" These 4 blocks in the Shortcuts app let you connect Siri with a @LangChainAI app API. Learn how to use Siri for chatting with documents in 100+ languages using @CohereAI 's multilingual model & @FlowiseAI /LangChain in this walkthrough! Video link below.🧵 ↴

Tweet media one

7

27

165

@MisbahSy

Misbah Syed

1 year

For beginners wanting to learn @LangChainAI and build AI apps, hope this video gets you started on your learning journey. Covers topics of calling LLMs, PromptTemplate, Chains, Memory, and more using Python-based LangChain. Link to video: Repo link 🧵↓

Tweet media one

4

32

163

@MisbahSy

Misbah Syed

1 month

Not much, just a @streamlit app that takes a YouTube URL and generates viral screenshot posts. You might have seen those posts that quote a person on a given topic. Thought of making something quick as a simple experiment using @LangChainAI & @GroqInc

3

23

162

@MisbahSy

Misbah Syed

1 year

📢 Calling all aspiring AI builders!🚀 I'm thrilled to announce the pre-launch of my course, "Build AI Apps with No-Code" using @LangChainAI . Join me on this exciting journey to learn how to create incredible LangChain apps without any coding experience. Join using link below↓

Tweet media one

14

26

150

@MisbahSy

Misbah Syed

5 months

With a bit of tweaking, now we have a better version of the Youtube-to-article generator: - a nice intro to the article - clickable links in the ToC & in the article - proper heading/subheading structure - code blocks & image embeds What other videos to generate articles for?

@MisbahSy

Misbah Syed

5 months

An attempt to solve the LLM challenge by @karpathy Built an LLM-based app using @LangChainAI to go from a YouTube link to a nicely written markdown article. (still a work in progress)

18

69

709

9

20

153

@MisbahSy

Misbah Syed

1 year

In case you missed them, three important components of @LangChainAI - Prompts, Chains, and Memory - are explained with visuals in the tweets below. Will be sharing more visuals in the upcoming days.

5

25

153

@MisbahSy

Misbah Syed

1 year

Planning to launch a few @LangChainAI apps in the upcoming weeks. Experimenting with a Newsletter Agent, which brings the latest news, research, tools, and topics in a given domain. The first stage is to aggregate info based on the top 15 established newsletters in that domain.

Tweet media one

8

15

149

@MisbahSy

Misbah Syed

9 months

🎉🎉 Launching LangBotz - A marketplace to explore no-code AI Agents & Assistants. LangBotz benefits from @LangChainAI -based Agents and Assistants that can use any combo of LLMs, embedding models, and much more. Essentially, it's an open-source version of GPT Store.

4

22

146

@MisbahSy

Misbah Syed

1 year

Want to create @LangChainAI apps without coding? No problem! Join us in our latest video tutorial where we'll guide you through building a chat-with-PDF app. Upload docs, filter searches, and embed chatbots effortlessly. Built using @FlowiseAI @bubble and @render

Tweet media one

11

29

146

@MisbahSy

Misbah Syed

1 year

Vancouver's first AI x @LangChainAI meetup was a big success! 🔥🎉 We showcased awesome demos and presentations from the community. For those who couldn't join, here are tutorials, demos, and code links to catch up, in the order presented. 🦜🔗 ↓

Tweet media one

5

14

136

@MisbahSy

Misbah Syed

1 year

I challenged myself to explain the core concepts of @LangChainAI to an average #ChatGPT user, covering topics like chains, prompt templates, retrievers, agents, tools, memory, and more. I hope this video serves as a good starting point for your learning journey. Link below 🧵↴

Tweet media one

9

16

135

@MisbahSy

Misbah Syed

1 year

How would Retrieval change with increased LLM token limits? Do we need vectorDBs? That's what triggered this experiment. 🧵↴ For a PDF-search app, what if we send the entire text content to the LLM and ask questions based on that. Our usual retrieval chain would modify as such:

Tweet media one

9

22

132

@MisbahSy

Misbah Syed

11 months

A few ways you can benefit from the @LangChainAI Hub. 🧵↓ 1 - Benefit from the top prompts shared by the community. Filter based on likes, views, downloads, and tags. These prompts can be used anywhere, from a simple translation app to prompts powering RAG and agent apps.

4

25

132

@MisbahSy

Misbah Syed

1 year

Well, now you can chat with the recommendation algorithm released by @Twitter @elonmusk at Using @LangChainAI @autodoc_ and template repo by @mayowaoshin .

7

26

128

@MisbahSy

Misbah Syed

1 year

In case you would like to use Guanaco by @Tim_Dettmers in @LangChainAI as part of an LLM chain, this code might come in handy. 😉 I am guessing, the chat model might also be coming soon to LangChain.

Tweet media one

@Tim_Dettmers

Tim Dettmers

1 year

QLoRA: 4-bit finetuning of LLMs is here! With it comes Guanaco, a chatbot on a single GPU, achieving 99% ChatGPT performance on the Vicuna benchmark: Paper: Code+Demo: Samples: Colab:

Tweet media one

90

948

4K

3

25

125

@MisbahSy

Misbah Syed

1 year

Connecting Siri to @LangChainAI app in 3 simple steps:

8

22

118

@MisbahSy

Misbah Syed

1 year

Build $$$-earning AI apps, not just Twitter demos. Finally! The course launch is complete. You can now build and deploy no-code @LangChainAI -powered SaaS apps. With over 60+ lessons & 14+ hours of content, create a portfolio app, LLM-chat app, Chat-doc app, Agent app, and more.

Tweet media one

3

19

94

@MisbahSy

Misbah Syed

10 months

Had this on my plate for a really long time... 🎉🎉 Introducing LangSaaS - now you can build & deploy your own chat document SaaS apps without code in no time! A @bubble SaaS template to build @LangChainAI doc chat apps. 🧵↓ 72 hrs launch sale!

Tweet media one

4

18

98

@MisbahSy

Misbah Syed

1 year

Detailed walkthrough with step-by-step instructions at DM for any questions. RT if you like the work. 😉

Tweet card media

Building a LangChain Agent (code-free!) Using Bubble and Flowise

Exciting news! We're launching a comprehensive course that provides a step-by-step walkthrough of Bubble, LangChain, Flowise, and LangFlow. Join us to learn ...

www.youtube.com

5

10

85

@MisbahSy

Misbah Syed

22 days

Anthropic has made it easy to share artifacts. Thought of aggregating top artifacts in one place to make it easy to discover all community-shared artifacts. Introducing , a free tool to discover, submit, and upvote cool Claude artifacts. A simple app

@alexalbert__

Alex Albert

24 days

You can now publish, share, and remix artifacts. Let's see what the coolest artifacts people have made are. Reply with your favorites below.

58

74

821

3

15

83

@MisbahSy

Misbah Syed

1 year

Which language should we use when responding to users? Can it be extracted from a user query? That's what we experiment with here. Benefiting from the 10 million Wikipedia vectors embedded by @CohereAI and hosted by @weaviate_io , built a QA system adding @LangChainAI parsers 🧵⇓

6

10

74

@MisbahSy

Misbah Syed

6 months

Fiction books to AI Movie generation: A rough blueprint of how fiction books can be made into movies using AI in the near future using tools available by @LangChainAI @UnstructuredIO @elevenlabsio @replicate @midjourney Maybe I could try building a quick MVP, let’s see.

Tweet media one

6

9

74

@MisbahSy

Misbah Syed

5 months

Working with @GroqInc 's API within @LangChainAI is a lot of fun. Everything runs at lightning speed. ⚡️⚡️ A few examples and code snippets below to get you started. 1/ Chat LLM Calls - Ofc, these are single-line replacements with a significant speed increase. For the provided

7

15

74

@MisbahSy

Misbah Syed

1 year

Feel free to experiment further with the open-sourced repo at Follow & RT to appreciate the work. Feel free to DM me for any help building a LangChain app.

Tweet card media

GitHub - menloparklab/cohere-weaviate-wikipedia-retrieval: A backend API to perform search over...

A backend API to perform search over Wikipedia using LangChain, Cohere and Weaviate - menloparklab/cohere-weaviate-wikipedia-retrieval

2

8

70

@MisbahSy

Misbah Syed

1 year

Hey Vancouverites! I am planning to host a LangChain/LLM in-person event in the Vancouver area. Would love to bring together @LangChainAI /LLM enthusiasts and build some cool stuff together. Let me know if this is of interest by filling out the form.

Tweet card media

LangChain/Large Language Model(LLM) IRL Event Pre-Registration - Vancouver

Event Timing: TBD (Most likely 1st weekend of May) Event Address: TBD (Greater Vancouver Area) Contact at [email protected]

docs.google.com

3

11

69

@MisbahSy

Misbah Syed

1 year

This code is also open-sourced at Deploy as a backend, and call it from any no-code/low-code or full stack frontend. Contact us @MenloParkLabAI for any help with building LangChain Apps.

GitHub - menloparklab/langchain-cohere-qdrant-doc-retrieval: This Flask backend API takes a...

This Flask backend API takes a document in multiple formats and allows you to perform semantic search using Langchain, Cohere and Qdrant. - menloparklab/langchain-cohere-qdrant-doc-retrieval

5

5

65

@MisbahSy

Misbah Syed

2 months

Want a month's worth of blog posts in MINUTES? Introducing Blogger Agent Team! Just give it a product name & URL & our AI agents will: 1️⃣ Extract features 2️⃣ Craft use cases 3️⃣ Write blog posts for EACH! ✍️ See the future of content creation!

Tweet media one

4

15

63

@MisbahSy

Misbah Syed

1 year

Now, anyone can build a full chat-data application like Chatbase or PDFai using a complete no-code framework. The biggest module of the course on building a chat-with-data app using no-code @LangChainAI dropped yesterday. We start from a wireframe:

Tweet media one

Tweet media two

Tweet media three

2

7

60

@MisbahSy

Misbah Syed

1 year

Since there is a lot going on with templates, thought of sharing a template as well. With this repo, now you can run @mckaywrigley 's ChatbotUI with a Python-based @LangChainAI app. Shoutout to @ajndkr for major contributions and for adding @lanarky_io for streaming. Link below ↓

Tweet media one

1

12

53

@MisbahSy

Misbah Syed

2 months

🎉 Thrilled to announce that I'm joining the @langflow_ai team! It's been an exciting journey collaborating with the LangFlow team since the early days of LangFlow (screenshot below from a webinar together on no-code LangChain), and now I'm officially joining the team.

Tweet media one

17

6

53

@MisbahSy

Misbah Syed

1 year

Certainly an honor to be mentioned by @hwchase17 . I don’t think I have seen a founder as helping and down to earth as Harrison. Rooting for his success.

@hwchase17

Harrison Chase

1 year

🎉🎉 @MisbahSy won our “chat-your-data” challenge a few months ago (and has won a few other hackathons in past months) Definitely someone to follow in this space

3

4

40

0

2

48

@MisbahSy

Misbah Syed

1 year

Learn with a step-by-step walkthrough on how to deploy PrivateGPT as an app: Reach out to us @MenloParkLabAI , we can help build LangChain apps, with your data privacy in mind. Or DM me. ;)

Tweet card media

Easiest way to deploy privateGPT app! Ideal for Businesses & Organi...

Exciting news! We're launching a comprehensive course that provides a step-by-step walkthrough of Bubble, LangChain, Flowise, and LangFlow. Join us to learn ...

www.youtube.com

5

7

50

@MisbahSy

Misbah Syed

1 year

@marktenenholtz Some of the best explainers are by @JayAlammar For example-

1

0

51

@MisbahSy

Misbah Syed

1 year

Originally introduced by @ivanmartit , PrivateGPT uses a similar architecture to the chat-with-pdf apps shared previously, except that @huggingface embedding model is used locally for vectorizing the text. And then relevant results are sent to GPT4ALL for response generation.

Tweet media one

2

2

49

@MisbahSy

Misbah Syed

11 months

The Udemy course on no-code @LangChainAI is free for all to join, for a limited time. We dive deep into LangChain concepts and use @FlowiseAI and @logspace_ai 's LangFlow for interactive learning. Feel free to share.

@LangChainAI

LangChain

11 months

. @MisbahSy is an incredible teacher and user of langchain. This class should be at the top of your list!

1

8

46

4

13

49

@MisbahSy

Misbah Syed

2 months

This is BIG! We are going live with @GroqInc , @LangChainAI , & @langflow_ai next week! See how Groq's fastest AI inference technology integrates seamlessly with LangChain (code) & LangFlow (no-code) Save your spot:

Tweet card media

Build Blazing-Fast LLM Apps with Groq, LangFlow, & LangChain

Register now for Build Blazing-Fast LLM Apps with Groq, LangFlow, & LangChain on crowdcast, scheduled to go live on June 12, 2024, 09:30 AM PDT.

www.crowdcast.io

2

11

47

@MisbahSy

Misbah Syed

5 months

Automated optimization is not yet part of RAGTune, but to be added soon. Open-source repo link: Feedback/suggestions are welcome. ;)

GitHub - misbahsy/RAGTune: Tuning and Evaluation of RAG pipeline. (Automated optimization to be...

Tuning and Evaluation of RAG pipeline. (Automated optimization to be added soon) - misbahsy/RAGTune

4

4

46

@MisbahSy

Misbah Syed

1 year

Since some folks reached out asking for a walkthrough, I made a video of how the backend could be deployed on @render with a few clicks, connecting with a @bubble frontend. @LangChainAI backend handles calls to @CohereAI , @qdrant_engine & @OpenAI . Link:

Tweet media one

@MisbahSy

Misbah Syed

1 year

With these few lines of code, you can deploy a backend for multilingual semantic search over a PDF using @CohereAI and @qdrant_engine , with the ability to embed and perform a search in 100+ languages. Certainly, you can see the power of @LangChainAI here.

Tweet media one

7

55

333

8

9

45

@MisbahSy

Misbah Syed

1 year

🎉🎉 Excited to share that my Udemy course on @LangChainAI is now launched. And below is how you can get it for free. It contains visuals and practical examples of how to use various LangChain components. - Prompts:

Tweet media one

@MisbahSy

Misbah Syed

1 year

Prompt Templates can be quite powerful if used the right way. Let's explore three types of prompt templates you can use in @LangChainAI . 1) Basic prompt template - Uses a single input from the user and combines it with the template string to send the combined prompt to the LLM.

14

58

283

5

10

42

@MisbahSy

Misbah Syed

11 months

The app is built using @streamlit and hosted on Streamlit cloud. It currently includes 7 lessons and two different styles of teaching. Link to the app:

LangChain: Getting Started Class

app screenshot

langchain-teacher.streamlit.app

3

3

43

@MisbahSy

Misbah Syed

1 year

Launching - get instant feedback on your resume, and generate a compelling LinkedIn headline and summary in seconds. The tool can also help you create customized cover letters based on specific job descriptions. Using @LangChainAI @bubble @render @OpenAI 🧵

2

10

43

@MisbahSy

Misbah Syed

1 year

FastAPI backend could be separately deployed and connected with any frontend framework. Works for privacy-centric document retrieval. Will make a vid soon to showcase the deployment steps. 😉 Repo open-sourced at Thanks @ivanmartit for privateGPT.

Tweet card media

GitHub - menloparklab/privateGPT-app: A fastAPI backend and a streamlit UI for privateGPT. Interact...

A fastAPI backend and a streamlit UI for privateGPT. Interact privately with your documents as a webapp using the power of GPT, 100% privately, no data leaks - menloparklab/privateGPT-app

3

6

41

@MisbahSy

Misbah Syed

1 year

Sharing the open-source chatbot using @LangChainAI and @huggingface repo with the PH community today.

Tweet card media

Chat Your Data - Product Information, Latest Updates, and Reviews 2024 | Product Hunt

Are you weary of incurring expenses for OpenAI LLMs? If so, why not try Langchain and models hosted on HuggingFace to construct your own ChatGPT-style chatbot with your own dataset? Star this...

www.producthunt.com

1

5

38

@MisbahSy

Misbah Syed

1 year

Excited to share that won the @CohereAI x @qdrant_engine hackathon as the top project. Thanks a lot, @lablabai for organizing the event. And thanks @CohereAI x @qdrant_engine for the great platform, and for sponsoring the event. 🙏🙏

@lablabai

lablab.ai

1 year

The @CohereAI and @qdrant_engine Multilingual Semantic Search Hackathon is finished! And let us tell you, it was one heck of a ride! 😎 🏆 The winner is AI Brand Intel And the finalists: Legal ease, Joan Holloway, EasyFind & Logos - the wild card, chosen by voting❤️

0

4

22

4

8

37

@MisbahSy

Misbah Syed

11 months

For folks starting to build using @LangChainAI , the “Use Cases” section has a wealth of information with well written code notebooks. Chances are that you might find your use case or something similar already covered in here.

0

10

34

@MisbahSy

Misbah Syed

4 months

🎉Introducing Flowise Streaming Chat *Plugin* for Bubble! Now you can stream responses from your @LangChainAI or @llama_index LLM apps within @FlowiseAI to a Bubble app and increase the chat response speed significantly. No more waiting for response completion!

1

2

34

@MisbahSy

Misbah Syed

6 months

Now, LangSaaS template is available in Python+JS. It's a RAG SaaS template built on Python @LangChainAI and @nextjs frontend. The template currently uses @qdrant_engine as a vectorDB and @cohere multilingual embedding model. It's integrated with @stripe for payment. 🔗↴

2

4

31

@MisbahSy

Misbah Syed

1 year

Launching , a brand monitoring solution that understands and responds in over 100 languages. You can scan social, and news mentions, generate AI responses based on your brand policies, and more. Built using @LangChainAI @bubble @qdrant_engine @CohereAI 🧵

4

6

33

@MisbahSy

Misbah Syed

5 months

It currently uses multiple LLM agents for various app steps. This is my workflow so far. Audio portion was easy, whisper it and get the timestamps/transcript. Video frames are a bit tricky for deciding to either embed it directly, ignore it, or take the content from the image

Tweet media one

8

2

33

@MisbahSy

Misbah Syed

2 months

For folks who prefer reading along @karpathy 's new video, converted the video to a written version here:

@karpathy

Andrej Karpathy

2 months

📽️ New 4 hour (lol) video lecture on YouTube: "Let’s reproduce GPT-2 (124M)" The video ended up so long because it is... comprehensive: we start with empty file and end up with a GPT-2 (124M) model: - first we build the GPT-2 network - then we optimize

Tweet media one

442

2K

16K

1

4

32

@MisbahSy

Misbah Syed

5 months

It takes a PDF and first extracts all the visual elements using @UnstructuredIO (they make it super easy). Additionally, each page of the PDF is saved as an image. Visuals and images are then sent to a vision model. Both GPT and Cluade were tested, and both performed well at it.

Tweet media one

3

4

28

@MisbahSy

Misbah Syed

3 months

📣 70+ languages are now supported on Brainy Docs! How? Prompt templates in @LangChainAI come in quite handy. Plug in the input/output language, and now content can be generated in 70+ languages.

3

5

27

@MisbahSy

Misbah Syed

10 months

Wow! This definitely makes it easy to deploy Langchain Apps. Now I don’t need to write custom api endpoints. Will be using this a lot. Great work @LangChainAI Team!

@LangChainAI

LangChain

10 months

🏓Introduction LangServe The best way to deploy your LangChains 📤Input/Output schema 📃/docs endpoint 🔠invoke/batch/stream endpoints 🎏/stream_log endpoint for streaming intermediate steps 🛠️LangSmith Integration Used to power ChatLangChain and WebLangChain Blog post and 🧵

Tweet media one

Tweet media two

Tweet media three

15

107

528

1

6

27

@MisbahSy

Misbah Syed

1 year

Full tutorial here: Let me know, what other topics would you like to see covered in future tutorial videos.

Tweet card media

Build a LangChain-based Semantic PDF Search App with No-Code Tools...

Exciting news! We're launching a comprehensive course that provides a step-by-step walkthrough of Bubble, LangChain, Flowise, and LangFlow. Join us to learn ...

www.youtube.com

3

8

28

@MisbahSy

Misbah Syed

2 months

Video content repurposing tools are usually quite expensive. But using this trick, you can repurpose video content completely FREE. From a YouTube URL, we can now generate multiple video clips, and even a title & a description for each clip. Complete no-code flow below ↓

2

5

27

@MisbahSy

Misbah Syed

10 months

LangSaaS is live on Product Hunt today, don't forget to smash that upvote button: 💸 Last day to get half off the launch price! 💵 And over 30 builders already received their copies of LangSaaS and will be building no-code @LangChainAI SaaS websites.

Tweet card media

LangSaaS - No Code LangChain SaaS - Product Information, Latest Updates, and Reviews 2024 | Product...

A Bubble SaaS template to build LangChain document chat apps. With pre-built integrations to Stripe, Flowise, Pinecone, Cohere, and unstructured.io, LangSaaS takes away days of development and...

www.producthunt.com

@MisbahSy

Misbah Syed

10 months

Had this on my plate for a really long time... 🎉🎉 Introducing LangSaaS - now you can build & deploy your own chat document SaaS apps without code in no time! A @bubble SaaS template to build @LangChainAI doc chat apps. 🧵↓ 72 hrs launch sale!

Tweet media one

4

18

98

0

7

26

@MisbahSy

Misbah Syed

1 year

That's where Contextual Compression in LangChain comes into play. Using @CohereAI 's rerank model available in LangChain, we can rerank results and select the top few most relevant results for the next stage.

Tweet media one

2

1

26

@MisbahSy

Misbah Syed

1 year

But thanks to @CohereAI and @weaviate_io , this was already done for us. Now the next step is to retrieve it using RetrievalQA chain in LangChain and the Python-based implementation as shown earlier.

Tweet media one

2

1

26

@MisbahSy

Misbah Syed

1 year

7 Ways to Use LangSmith's Superpowers to Turboboost Your LLM Apps Additionally, includes step-by-step instructions on how you can use LangSmith with Python-based @LangChainAI , and no-code tools @FlowiseAI & @logspace_ai 's Langflow. Video Link below ↴

Tweet media one

2

10

25

@MisbahSy

Misbah Syed

1 year

Building a typical document search system begins with first extracting the text (Wikipedia in our case), splitting it into chunks, embedding them, and then saving it to a vector database.

Tweet media one

2

0

25

@MisbahSy

Misbah Syed

1 year

Bonus: Retrieval functions are also wrapped in python-flask API routes, you could have a @Wikipedia retrieval backend deployed in a few clicks. 😉

3

0

25

@MisbahSy

Misbah Syed

1 year

This is cool! @merlindotcom_ got an app built in a few hours for the Wikipedia embeddings made available by @CohereAI and @weaviate_io . Check it out, it's pretty neat.

@merlindotcom_

Michael Houghton

1 year

Built another App with @streamlit . Introducing WikiGPT 📖 Powered by @CohereAI , @weaviate_io , and @LangChainAI , this app grants access to 10 million Wikipedia embeddings in 10 languages. Shout out to @MisbahSy for the base code!

10

14

99

2

3

24

@MisbahSy

Misbah Syed

7 months

Let’s do a LangChain x AI meetup/event in India! We can have demos, workshops, showcases, basically anything you have worked with @LangChainAI . Planning to visit in Jan, and would be awesome to meet some builders and enthusiasts there. Lemme know if interested👇

5

4

24

@MisbahSy

Misbah Syed

11 months

Made a YouTube video based on the tips shared yesterday on using @LangChainAI Hub. Excited to see what community shares and builds using this new feature. Link below ↓

Tweet media one

@MisbahSy

Misbah Syed

11 months

A few ways you can benefit from the @LangChainAI Hub. 🧵↓ 1 - Benefit from the top prompts shared by the community. Filter based on likes, views, downloads, and tags. These prompts can be used anywhere, from a simple translation app to prompts powering RAG and agent apps.

4

25

132

1

7

22

@MisbahSy

Misbah Syed

11 months

Hey Vancouverites, our second AI x @LangChainAI meetup will be on October 11th. Enjoy the presentations and network with other AI builders & Langchainers. Register at Limited space, don't miss out.

4

9

23

@MisbahSy

Misbah Syed

1 year

It was great presenting at the @LangChainAI webinar, alongside @FlowiseAI and @logspace_ai . Replay: I talked about building apps using @CohereAI x @qdrant_engine with @bubble At @MenloParkLabAI we can help you build LangChain apps quickly. Reach out!😉

Tweet card media

LangChain No-Code Webinar

Register now for LangChain No-Code Webinar on crowdcast, scheduled to go live on April 26, 2023, 09:00 AM PDT.

www.crowdcast.io

4

6

23

@MisbahSy

Misbah Syed

6 months

5 amazing AI automation made super easy by @AgentHub_AI . (completely no-code!) 1/ Inbox Summarizer: Reads all the unread emails in your inbox and generates a summary email for you using AI, outlining the emails worth reading.

Tweet media one

4

3

21

@MisbahSy

Misbah Syed

1 year

As shared previously, it basically divides a document retrieval system into two API functions corresponding to document ingestion and retrieval/search query.

Tweet media one

2

3

23

@MisbahSy

Misbah Syed

3 months

Day 1 update: We've already hit 100+ users on the app! Let's keep this momentum going! 🚀

Tweet media one

@MisbahSy

Misbah Syed

3 months

🚀 Introducing Brainy Documents! 🧠📄 Generate explainer videos for your PDFs using AI. Drop a PDF and let AI generate videos explaining the content for you. Built using @LangChainAI @DeepgramAI @UnstructuredIO BONUS: Brainy Documents can also generate presentation slides.

8

41

228

4

2

21

@MisbahSy

Misbah Syed

1 year

The code is open-source at Feel free to deploy this as a backend, and call it from any no-code/low-code or a full stack frontend.

Tweet card media

GitHub - menloparklab/langchain-cohere-qdrant-retrieval: This is a template retrieval repo to...

This is a template retrieval repo to create a Flask api server using LangChain with Cohere embeddings and Qdrant Vector Database - menloparklab/langchain-cohere-qdrant-retrieval

2

2

20

@MisbahSy

Misbah Syed

2 months

Each agent uses @GroqInc to call Llama3-70b in @langflow_ai . The agents are called from @n8n_io to take form submission and then write agent responses to Google Sheets. One could also directly post them to blogging websites.

1

1

22

@MisbahSy

Misbah Syed

11 months

LangChain Teacher app showcased amongst the favorites 🫶

@LangChainAI

LangChain

11 months

🖼️ Introducing Made With LangChain🖼️ A gallery of our favorite applications built with LangChain We’re so inspired by the creativity and ambition across our community and we think (and hope!) this site might inspire you, too! check it out at ✨

Tweet media one

11

28

204

2

3

21

@MisbahSy

Misbah Syed

1 year

Hey, look what I received. @LangChainAI merch sent by @hwchase17 . 🥳 🥳 I am planning to double down on LangChain content in the coming weeks. Let me know if any particular content you would like to see. Articles? Videos? Courses? Boilerplate code? Template repos? 🧵

Tweet media one

8

1

20

@MisbahSy

Misbah Syed

6 months

A walkthrough video on the thread. Learn LangGraph by @LangChainAI , the easy way. Link to YouTube video and the Notebook below.

@MisbahSy

Misbah Syed

6 months

Understand LangGraph by @LangChainAI , by building simple graph apps. 🧵↓ Let's start with a simple app with two nodes attached by an edge

Tweet media one

4

29

177

2

4

21

@MisbahSy

Misbah Syed

5 months

Article Generator step also took a few attempts to make sure the article was coherent and in the style intended. Happy to take your feedback on what aspects of the article writing to improve. The article can be viewed at:

Tokenization Video Conversion

This is an attempt at solving the LLM Challenge put out by Andrej Karpathy on X (twitter)

misbahsy.github.io

1

1

21

@MisbahSy

Misbah Syed

1 month

Will be hosting a live "Hands-on Workshop: Build an AI Web App with Bubble & Langflow" tomorrow. We'll build a flow in @langflow_ai , call @GroqInc API for LLMs, and then connect it with a @bubble -based web app. Completely no-code! Join us and build a web app together.

Tweet media one

2

4

21