AI Tech Suite
SciPhi's R2R (RAG to Riches) is an advanced AI retrieval system designed for production-ready Retrieval-Augmented Generation (RAG) applications. It helps developers bridge the prototyping and launching phases of robust AI applications by offering features like agentic RAG, ingestion, document management, hybrid search, and automatic knowledge graphs. R2R processes millions of documents in various formats (40+), provides granular access control, and builds knowledge graphs to enrich context. It boasts a developer-friendly API and integrates with various platforms like OpenAI, Vertex AI, and Anthropic.
Responsibilities:
Experimenting with new data augmentation strategies for improved relevance ranking. Building next-generation summarization and knowledge graph extraction pipelines. Designing a distributed retrieval pipeline that can handle billions of text chunks with sub-100ms latency. Building out new caching layers or microservices to reduce cost and improve reliability. Optimizing throughput for ingestion workflows. Qualifications:
Experience with container orchestration (e.g., Kubernetes, Docker) and distributed systems. Tackled large-scale production challenges—database sharding, queue orchestration, or microservice architecture. A systems thinker who loves optimizing at every level of the stack. A knack for building intuitive UIs and seamless user experiences. Blend of design thinking with solid engineering skills. Enjoy iterating quickly based on real user feedback.
#J-18808-Ljbffr
Experimenting with new data augmentation strategies for improved relevance ranking. Building next-generation summarization and knowledge graph extraction pipelines. Designing a distributed retrieval pipeline that can handle billions of text chunks with sub-100ms latency. Building out new caching layers or microservices to reduce cost and improve reliability. Optimizing throughput for ingestion workflows. Qualifications:
Experience with container orchestration (e.g., Kubernetes, Docker) and distributed systems. Tackled large-scale production challenges—database sharding, queue orchestration, or microservice architecture. A systems thinker who loves optimizing at every level of the stack. A knack for building intuitive UIs and seamless user experiences. Blend of design thinking with solid engineering skills. Enjoy iterating quickly based on real user feedback.
#J-18808-Ljbffr