SciPhi is a cloud platform for developers that simplifies building and deploying serverless RAG pipelines. Built on top of the open source R2R framework, it enables builders to focus on creating innovative applications rather than managing infra.
Exciting news, Product Hunters! ๐
I'm thrilled to introduce SciPhi โ the ultimate cloud platform for Retrieval Augmented Generation (RAG). SciPhi empowers developers to effortlessly build and deploy serverless RAG pipelines with just one click!
Built on top of the open-source R2R framework, SciPhi streamlines the entire process, allowing developers to focus on what truly matters โ creating innovative applications that push the boundaries of what's possible with AI.
Our early users have reported that SciPhi+R2R's streamlined workflow saves them significant time and eliminates the headaches associated with deploying to production and iterating on their RAG pipelines.
With SciPhi, you can:
โ Deploy production-ready RAG pipelines in seconds
โ Customize your pipeline using intuitive configuration files
โ Extend your pipeline logic with code
โ Secure your sensitive data with encrypted secret management
โ Automatically scale your pipeline as your application grows
โ Monitor and optimize your system with comprehensive insights
We've been working tirelessly to bring you a platform that accelerates the development, deployment, and optimization of RAG pipelines, all without the hassle of managing infrastructure. Whether you're a seasoned developer or just starting your AI journey, SciPhi provides a seamless experience tailored to your needs.
But don't just take our word for it โ dive into our comprehensive documentation and see for yourself how easy it is to get your RAG solution up and running.
Join our vibrant community of developers, researchers, and enthusiasts on Discord, where you can connect with like-minded individuals, get support, and stay up-to-date with the latest advancements in the RAG ecosystem.
We can't wait to see the incredible applications you'll build with SciPhi! Share your thoughts, ask questions, and let us know how we can help you make the best RAG pipeline.
Happy building! ๐ ๏ธ
@emrgnt_cmplxty Congrats on the launch! I actually worked on something similar last year - an open-source deploy RAG to your own cloud tool called RAGStack.
The problem we ran into was that while hackers were interested in messing around with it, we couldn't find developers at actual companies that needed a solution to deploy RAG. AI startups viewed it as a core competency and larger companies had teams of ML engineers to work on the problem.
What does your ideal customer profile look like at SciPhi?
@ayan_bandyopadhyay It's awesome that you worked on something similar, thanks for sharing your experience.
I've heard about the graveyard of companies that could not get traction in providing RAG solutions. We are betting that as RAG becomes more and more ubiquitous the market for easier solutions is going to grow rapidly.
Right now, our ICP is a developer that needs RAG in production, but who has numerous other concerns around building out their application. They will use SciPhi to get a solution out quickly, and that solution is one that they can customize and iterate on. We are very focused on making sure the developer has full control over selected providers and integrations.
E.g. our bet is one that RAG will become boilerplate for many uses cases over time - like so many similar services have (web hosting, authentication, database management).
Congrats on the launch @emrgnt_cmplxty ๐
Does SciPhi support other customization options beyond configuration files for flexibility e.g. script files?
All the best with the launch Owen
@jgani - To answer your question, yes!
I recommend taking a look at the docs core-features, starting here [https://r2r-docs.sciphi.ai/core-...].
Also, there is a SciPhi assistant that can be toggled in the bottom right of the docs / web app that can answer questions such as this.
The reply it gave to your prompt above was -
`
SciPhi provides customization primarily through the config.json file, where you can specify various settings including vector database providers, embedding settings, language model providers, and more. This file serves as the central point for tailoring your RAG pipeline to your specific needs.
Additionally, SciPhi offers a range of RAG pipeline templates that serve as starting points for building RAG pipelines with different capabilities and configurations, which include customization options for vector database provider, evaluation settings, embedding settings, text splitter settings, and language model provider.
...
`
Report
"Hey there, fellow trader! ๐ Have you heard of Wini_fx_team on instagram ? It's a great platform that makes trading fun and easy. With a wide range of assets, low fees, and a user-friendly interface, it has everything you need to succeed. And the best part? It's totally free to sign up! What are you waiting for?
Report
SciPhi is pretty sick. Was able to configure and deploy a RAG to talk to some really dense PDF documents and an easy pipeline to to it, all in just a few minutes. That would've taken me a few days to do in LangChain, not to mention the deployment!
@andrewwang97 - Thanks Andrew - it was exciting to watch SciPhi accelerate your build time.
Report
Congrats on the launch! This is so cool. Really interested in seeing how this will grow to support ingestion from more complex documents especially as well.
@raunak_chowdhuri Thanks, right now we have a simple implementation and an integration with Reducto.ai. This will expand as we get more feedback on customer needs.
Congrats, Owen! Development acceleration efforts are always so valuable, good luck to you and your team! As the founder of a no-code website builder, we'll see if SciPhi comes in handy for our team. In the meantime, if you'd like to wrap your expertise in an eLearning product, you're welcome to check out our course builder ๐
R2R
Dealwise
R2R
Dealwise
Unofficial Product Hunt Chrome Plugin
R2R
R2R
R2R
Tilda
R2R