Skip to content

Latest commit

 

History

982 Commits

Folders and files

NameName
Last commit message
Last commit date
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

SproutOS

SproutOS enables users to cost-effectively create and deploy database-backed workflows and backend-supported, auth-backed sites without knowing how to code.

Features

App Store

We host an app store of open source apps and workflows. At the moment, this includes an Android play store and a website store. Users can customize open source apps for personalization needs, and our infrastructure will keep users' forks up-to-date, folding upstream fixes and features into the personalized codebase.

We want to democratize creating personalized apps and workflows. An open app store enables this by making the large hurdle of creating the initial app already done through our ecosystem and personalization is only a sentence in a coding agent away.

Fork maintenance can use your existing Claude Code subscription or an API key to a service including in-house LLM hosts.

Serverless Backend Operations

We host serverless backend operations including:

  • Serverless sites: we recreated Vercel Fluid Compute, hosting websites for a few cents per month
  • Workflows: cron and on-demand based workflows and background jobs spun up only when used. We use Firecracker to execute your custom code, and use a managed service to intelligently deduce when and for how long to have your workflow code running.
  • Databases: serverless Postgres allows us to bill few cents per month with fast cold starts.
  • Tenant-split databases: We host a shared Valkey and Elasticsearch service that is tenant-split
  • Discounted AI token prices: through our service, we can reduce your AI token cost through our service-wide purchasing of API tokens from LLM providers

Because users can spin up databases cheaply, this means an app that values users' privacy and data ownership can use OAuth Login to SproutOS and create a database dedicated to the user, allowing data ownership without needing a fork. Data ownership and personalization are core tenants of SproutOS. A user doesn't need to fork to own their data either if the original app enables personalized databases.

Data ownership enables users to use coding agents to transfer their data to other apps in a similar vertical; data ownership reduces switching costs and can help flourish diversity in app ecosystems.

Sites and Workflows

We also host sites, like Lovable and OpenAI Sites, and workflows, like n8n. The difference is we ensure code is open source and our infrastructure does not lock you in. For privacy, personalization, and data ownership focused consumers, you'll love our fork maintenance and app stores.

Sprout sites are also useful for setting up RAG pipelines. For example, iMessage has a horrible search result reputation. Sprout's community coding agent skills learned from top Silicon Valley startups how to set up customized, refined, and fine tuned RAG that utilize more than just the basic embedding models and embedding databases. With those skills, Sprout can create a website for searching messages and set up the necessary backend services for cheap and provide better results.

For Businesses

For businesses and AI consultancies, our service allows employees to conveniently create automations with just a few sentences. For example, customer success managers are overworked and wish they could program custom follow-up messages to clients or cancel if they followed up already. For different departments, SproutOS Sites enable employees of different departments to create their own RAG pipelines. Users would simply need to connect with Sprout/Claude's coding agent, integrate with their chosen data source, tell the coding agent what their expected search queries and results should look like, and utilize Sprout's skills to create a custom RAG pipeline for one's use case. Skills go beyond embeddings, using fine tuned embeddings and coding agent skills from our community that can build out multiple backend services like a knowledge graph and refine what data to ingest. See this ByteByteGo article for different examples of use case based RAG pipelines.

We remove the need for businesses to provide technical resources to deploy their ideal automations and websites. Sprout's backend operation solutions reduce bureaucracy by empowering employees to deploy whatever software they need for super cheap.


Scaffolded from Andrew-Chen-Wang/nextjs-spa-split. Licensed BSD 3-Clause — see LICENSE.

Sponsor this project

Used by

Contributors

Languages