An LLM-powered, conversational search and analytics platform.
Use Sycamore to answer questions on your complex unstructured data. Customize RAG pipelines, and go beyond RAG with analytics.
AI-powered segmentation, enrichment, and structuring tools to prepare your data for high-quality results.
Built on OpenSearch for scalable vector indexing and enterprise-grade security.
Secure and scalable
100% open source, Apache v2.0
To get the right answers, simple RAG is not enough. When data gets complicated, developers reach for Sycamore.
One-stop Shop for Developers
Everything you need to build conversational apps on complex unstructured data, from data crawling to preparation to indexing to conversational APIs. Iterate and get to production faster, and scale up seamlessly as your dataset grows.
Robust and scalable API to construct Python-native data pipelines to clean, segment, enrich, and structure complex data in PDFs, HTML, tables, and graphics. Leverage LLMs to extract and manipulate data, and generate vector embeddings.
High-quality Search and Analytics
Higher-quality answers with APIs for conversational RAG, hybrid search (vector and keyword), and analytics on structured metadata (OpenSearch API compatible). You can also use post-processors like re-ranking and near-duplicate detection.
Open Source and Customizable
100% open source (Apache License v2.0), with the freedom to customize and build. Built with OpenSearch, an enterprise grade search engine. Easily experiment with different generative AI models and customize search pipelines.
Versatile to handle the complexities of
Deploy Sycamore with Docker and choose your hardware. Use default data preparation code or pipelines that are customizable with a built-in Jupyter notebook. Choose your LLMs and data processing AI models, ingest your data, and query it with natural language or analytics. Sycamore uses a scalable, secure, and customizable OpenSearch backend for indexing and data retrieval.
Scale up quickly
Sycamore demo videos:
See how you can harness your unstructured data with Sycamore.
Developers use Sycamore as a core building block for search and analytics applications in financial services, healthcare, manufacturing, eCommerce, and customer support.
Research and discovery
Build apps to enable analysts and researchers to ask hard questions on complex unstructured data that includes tables, infographics, and complicated layouts. Discover and use critical information that would otherwise be missed.
Reporting on unstructured data feeds
Create structured reports from unstructured data to answer key business questions. Run scheduled pipelines that extract, enrich, and store information from diverse datasets, such as Salesforce data, health records, or contracts.
Technical knowledge bases
Empower technical knowledge workers with AI-assistants for manuals, technical documents, installation guides, and catalogs to quickly answer technical questions and find information. Get direct answers and pointers to sources.
Create co-pilots to empower customer support teams, healthcare professionals, or empower customers to directly query knowledge bases, support tickets, FAQs, healthcare records, and other information sources.
Aryn's mission is to answer questions from all of your data. Our passion is data and AI in the cloud. With the recent flurry of advances in AI, we believe that seamlessly processing and querying unstructured data is within reach. Towards this, we founded Aryn and built Sycamore to re-imagine enterprise search.
Up to 90% of enterprise data generated is unstructured, and requires search technologies to make it useful. Today's options require significant tuning, complex data preparation, and rummaging through result sets. Instead, what if you could just ask questions of your data, and easily get accurate and crisp answers?
We use LLMs as assistants to enhance and simplify the data pipelines that comprise search stacks. Also, we provide a data preparation and enrichment system for getting quality answers from LLMs and RAG pipelines. We handle the complexity of scaling, security, and manageability. And, it's all 100% open source.
Aryn (pronounced "air-in") means "high mountain." Over 90% of data today is unstructured, leaving enterprises with mountains of data that are difficult to conquer. Our goal is to help you summit that peak.
Aryn's team includes leaders with decades of big data, AI, and cloud experience from AWS, Google Cloud, Stripe, Dremio, HP, IBM, Yahoo!, and Meta.
Mehul is the CEO and co-founder of Aryn. Most recently, he was an executive at AWS and Google Cloud. At AWS, he helped create and launch OpenSearch as well as found and grow AWS Glue and Lake Formation. Prior to that, he was founder and CEO of Amiato, a cloud ETL service (acquired). He was also a scientist at HP Labs where he published in top-tier database and systems conferences and won several awards. He earned his PhD in databases from U.C. Berkeley, and degrees in CS and Physics from MIT.
Ben is the CTO and co-founder of Aryn. He has worked with data for his entire career and was most recently a principal engineer at AWS, where he spent time working on critical components of Amazon Redshift and helped found and scale AWS Glue and Lake Formation. Before that he was an early employee at Amiato, a cloud ETL service (acquired). He earned his PhD from Cornell University in the database group, and holds a computer science degree from Carleton College.
Jon is the founding Chief Product Officer at Aryn. Prior to that, he was the SVP of Product Management at Dremio, a data lake company. Earlier, Jon was a Director at AWS, and led product management for in-memory database services (ElastiCache and MemoryDB for Redis), Amazon EMR (Apache Spark and Hadoop), and founded and was GM of the blockchain division. Jon has an MBA from Stanford Graduate School of Business and a B.A. in Chemistry from Washington University in St. Louis.
Subscribe to our mailing list
Join our email list to stay up to date about Sycamore.