@openpj2Solution Architect, TAI Solutions
Shipped: OpenCrawling
OpenCrawling is a high-performance enterprise data ingestion and security mapping framework. Powered by Java 25 and Spring Boot 4, it orchestrates secure pipelines using Apache Tika for text extraction, Apache Kafka for queueing, Apache Ozone & S3 for distributed payload storage, Redis for caching, and Ollama for embeddings while preserving document permissions (ACLs).
RAGSpring AIJava 25EmbeddingData FederationData LiberationPipelineRAG Pipeline
Comments1
Keeping each document's permissions in place as OpenCrawling pulls company files into AI search solves the worry that stalls a lot of bigger companies, which is an assistant showing an employee a salary sheet or contract they were never allowed to open. For the IT and security teams you'd be selling to, that promise is likely what turns a cautious pilot into a design partner.
1 comment
Keeping each document's permissions in place as OpenCrawling pulls company files into AI search solves the worry that stalls a lot of bigger companies, which is an assistant showing an employee a salary sheet or contract they were never allowed to open. For the IT and security teams you'd be selling to, that promise is likely what turns a cautious pilot into a design partner.