langchainjs VS Converter

*** all code examples are using LangChain TS on the main branch on September 22nd, 2023 (roughly version 0.0.153).

At the time of the prototype's development, the Langchain GitHub loader sent one request per file to fetch the repository sequentially, leading to prolonged download times. In our case about 2 minutes for the insights.opensauced.pizza repository. This issue was later resolved in hwchase17/langchainjs#2224, enabling parallel requests for faster retrieval.

Another llm orchestration library for js/ts

You will probably need to create a custom agent with custom tools to do what you want to do. Look at Langchain (seems like there is an open PR for Google calendar tools here: https://github.com/hwchase17/langchainjs/pull/1777). There are a lot of great integration examples on their website (including for vectorDB memory https://python.langchain.com/docs/modules/memory/how_to/vectorstore_retriever_memory)

You can checkout there docs (here)[https://js.langchain.com/docs/].

The strategy I tried, was to point scalablytyped, at langchainJS.

TL;DR the BANGER TWEET BRAINSToRMER 💥 🐦 🧠 is an open-source, fullstack React/Express/Postgres/Pinecone app that brainstorms new ideas and tweet drafts based on your own notes/ideas and the tweets of your favorite twitter users. This isn't a bot, but you can think of it rather as your personal twitter intern that monitors current twitter **trends**, keeps note of your **ideas**, helps you **brainstorm** new ones, and write **draft** tweets. It's your job to find and edit the best ideas before saving them to your personal notes database or tweeting them out from the app itself. it uses pg-boss cron jobs via Wasp, OpenAI, langChain, and Pinecone for the vector store https://github.com/vincanger/twitter-brainstorming-agent

It is not great. It has a lot of limitations, but can be used under certain conditions. https://github.com/hwchase17/langchainjs/pull/655

The longer story is that st-material-ui incorporates https://github.com/ScalablyTyped/Converter/pull/487 in order to get much, much cleaner API. You'll probably have seen the fake literal types, some rewriting from type unions to inheritance, things like that.

Given that you want interoperability with js, I'd start by playing with https://scalablytyped.org/, then, play with the scalajs-react demos (https://github.com/ScalablyTyped/ScalaJsReactDemos) and the slinky demos (https://github.com/ScalablyTyped/SlinkyDemos). There are some libraries that scalablytyped doesn't support pretty well but you can leverage https://github.com/nafg/scalajs-facades for those.

ScalaJS is awesome. Really solid and mature project, can totally recommend.

The only thing that can be annoying is when you want to have a typesafe interface and have to write a lot of adapters for javascript libraries.

Fortunately there is even a project that can make use of typescript interfaces for those libraries, so that you can use them from ScalaJS more or less automatically: https://scalablytyped.org/

I'm using it mostly for full-stack web development with ScalaJS (https://www.scala-js.org) in the frontend (https://outwatch.github.io/docs/readme.html) and in the backend with AWS lambdas.

The ecosystem is currently in the process of porting all the libraries to Scala 3. So if you're new to Scala, I'd recommend to start with Scala 2, which is rock-solid and already very powerful.

I never worked with SQLAlchemy. But on the scala database side, popular libraries are Doobie (https://tpolecat.github.io/doobie) and Quill (https://getquill.io). Keep in mind that these are for Scala on the JVM. On the ScalaJS side I'm using the javascript library pg. But I'd like to try if it works well with Prisma soon.

The nice thing about ScalaJS is, that you can use Javascript libraries. And if there are typescript facades, then you can transpile these to Scala and use them in a type safe way (https://scalablytyped.org).