ayanb9440

Looks like somebody forgot to update the gitignore lol

Yup that's right its Robotic Process Automation.

Based on the feedback in this thread we're going to be releasing an updated version that focuses more around tooling for the browser agents themselves as opposed to scaling/scheduling, so stay tuned for that!

This should be fixed now

We're in the middle of putting this together right now but it's going to be a wrapper around Google Secret Manager for those that don't want to set up a secrets manager themselves.

Depends on the use case. Lots of hospitals and banks use RPA to automate routine processes on their EHRs and systems of record, because these kinds of software typically don't have APIs available. Or if they do, they're very limited.

Playwright and other browser automation scripts are a much more powerful version of RPA but they do require some knowledge of code. But there are more and more developers every year and code just gets more powerful every year. So I think it's a good bet to make that browser automation in code will replace RPA altogether some day.

We do support sentry. Finic projects are poetry scripts so you can `poetry add` any observability library you need.

That's a great suggestion! Essentially a cron job to check for website changes before your automation runs and possibly breaks.

What does this check look like for you? Do you just diff the html to see if there are any changes?

There are quite a few open source YC startups at this point. Our understanding is that:

1. Developer tooling should be open source by default 2. Open source doesn't meaningfully affect revenue/scaling because developers that would use your self-hosted version would build in-house anyway.

If you want to use an agent for scraping/automation, you would need to supply it with auth credentials. So permission is required by default.

Looking at their docs, it seems that with Browserbase you would still have to deploy your Playwright script to a long-running job and manage the infra around that yourself.

Our approach is a bit different. With finic you just write the script. We handle the entire job deployment and scaling on our end.

We currently don't save the browser state after the run has completed but that's something we can definitely add as a feature. Could you elaborate on your use case? In which scenarios would it be better to split a run into multiple steps?

I spent some time building legacy software integrations while contracting for startups, so I wrote this guide for developers in the same boat. It covers user-permissioned authentication, the trade-offs between RPA and reverse-engineering APIs, and how to deploy your integration to production.

I see so you'd prefer the flow to be enter your ARR -> log in to confirm as opposed to the other way around? I'm guessing to reduce friction?

Thanks for the heads up! We'll have that fixed ASAP but in the meantime signing up with Google or email/password should work

Here is a fully self-hostable solution that connects to PDFs in your google drive folder: https://github.com/ai-sidekick/sidekick

Uses weaviate so that even the vectorstore can be self-hosted

Pre-populating with test credentials is a good idea!

Some examples use cases for Sidekick include:

- Customer support bots that answer questions based on your knowledgebase

- Internal tools for searching through documentation scattered across different SaaS tools

Yeah for sure! I'm guessing you want a Plaid-like flow where end users OAuth and enter in which folder/etc?

Nice, quite a feature set. Add a PR and ping us on slack, we'll merge it in

+1, here's the line in the codebase: https://github.com/ai-sidekick/sidekick/blob/main/sidekick-s...

We'll also add support for huggingface embeddings like MPNET and SBERT in the future!

Is Marqo a vector database? If so that sounds great, feel free to put up a PR!

An API integration is in our roadmap! Send over an email to founders@getsidekick.ai for more details.

Gpt says:

It means that the bot may not always provide the correct response to a question, even if the answer can be found in the support documents. The bot may not have access to all the information or may not be able to understand the context of the question.

Good question, let me ask gpt

Likely Weaviate to replace Pinecone and Haystack to replace OpenAI

I can definitely see myself using this in previous roles!

I haven't tried haystack actually, what has been your experience with it so far?

Currently sleuth only looks at public channels, although user-specific private channels is on the roadmap.

Thanks for the feedback! Currently looking into swapping out the OpenAI + Pinecone portion with fully open source and self hosted options such as Weaviate or Qdrant for the next version.

Very exciting! I used this team's previous product (DVC) at a research lab at Caltech. This looks like a very useful tool.