{"id":79568,"title":"SciPhi - Build, deploy, and optimize AI retrieval","tagline":"An open source platform that enables developers to build the best RAG system.","body":"**TL;DR**: [SciPhi](https://www.sciphi.ai/) is a **cloud platform** that simplifies the deployment and optimization of production-ready Retrieval-Augmented Generation (**RAG**) pipelines for developers. Built on top of the robust **open source** **R2R** framework, SciPhi enables builders to focus on creating innovative applications rather than managing infrastructure.\n\n[**SciPhi Cloud Demo**](https://www.youtube.com/watch?v=ANWAGCzGQUU\u0026ab_channel=redfauna101)\n\n**The Problem**\n\nRAG is fundamentally a search problem, and search at scale is hard. While deploying a basic RAG pipeline is relatively straightforward, scaling it to meet the diverse needs of a growing user base and/or dataset presents significant challenges. As indexed data grows, query times can slow down, leading to a diminished user experience. Further, different users will have varying complaints about system performance, depending on their specific use case and data set.\n\n**📦 SciPhi's Solution**\n\n* Seamless deployment of RAG pipelines to the cloud in just a few clicks\n* Customizable configurations via an intuitive config.json file to specify vector DB, embedding settings, LLMs, etc.\n* Secure secret management to encrypt and store API keys and other sensitive info\n* Automatic scaling to handle increased traffic and data volumes as your app grows\n* Built-in monitoring, logging, and A/B testing to gain insights and optimize performance \n\n![uploaded image](/media/?type=post\u0026id=79568\u0026key=user_uploads/372342/4fa047ca-3982-4c2d-b0f9-f1ea5d48c348)\n\n![uploaded image](/media/?type=post\u0026id=79568\u0026key=user_uploads/372342/0f4b026d-ba30-4526-ba9d-5d3019a76df2)\n\n![uploaded image](/media/?type=post\u0026id=79568\u0026key=user_uploads/372342/3208055b-d3f6-43b8-b8dc-394621376af4)\n\n![uploaded image](/media/?type=post\u0026id=79568\u0026key=user_uploads/372342/2f7fc1cf-3fb1-4909-a61c-e6e8814946a5)\n\n**Meet the Founder**\n\nSciPhi was founded by [Owen](https://www.linkedin.com/in/owencolegrove/) (CEO), who brings a unique perspective to the world of RAG. With a PhD in particle physics and experience creating quantitative research strategies at Citadel, Owen understands the challenges of working with large-scale data systems. As a second-time founder, he grew frustrated by the lack of tooling available to optimize RAG pipelines while building an open-source semantic search engine+agent. This experience started him on the path to creating SciPhi.\n\n**Our Ask**\n\n* 🚀 Sign up for early access at [app.sciphi.ai/login](https://app.sciphi.ai/login)\n* 💬 Join our [Discord](https://discord.gg/p6KqD2kjtB) community to connect with other RAG enthusiasts\n* [⭐](https://emojipedia.org/star) Check out the [R2R Github repository](https://github.com/SciPhi-AI/R2R/tree/main) and leave us a star\n* 🙌 Share this post with developers and teams looking to harness RAG in production\n\nWe're on a mission to empower developers to build state-of-the-art RAG pipelines. If that resonates, come build the future with us! Reach out to [founders@sciphi.ai](mailto:founders@sciphi.ai) to chat.","slug":"KhM-sciphi-build-deploy-and-optimize-ai-retrieval","created_at":"2024-03-21T14:18:21.767Z","updated_at":"2026-05-25T01:53:27.250Z","total_vote_count":89,"url":"https://www.ycombinator.com/launches/KhM-sciphi-build-deploy-and-optimize-ai-retrieval","share_image_url":"https://www.ycombinator.com/media/?type=post\u0026id=79568\u0026key=user_uploads/372342/4fa047ca-3982-4c2d-b0f9-f1ea5d48c348","company":{"id":29181,"name":"Event Horizon Labs","slug":"event-horizon-labs","url":"https://www.ehl.markets/","logo":"https://bookface-images.s3.amazonaws.com/small_logos/952b6381b227fe13913d082e7f9bdec4731e8adb.png","batch":"Winter 2024","industry":"Fintech","tags":["Finance","Investing","AI"],"search_path":"https://bookface.ycombinator.com/company/29181"}}