Oraion built an AI-native data platform using LangChain, LangGraph, and LangSmith to surface actionable insights from fragmented enterprise data. The stack reveals a company focused on LLM application patterns—RAG pipelines, agent automation, fine-tuning—rather than infrastructure. Active hiring across engineering, sales, and data suggests product-market validation is driving early-stage scaling, though hiring velocity remains cautious (2 roles in 30 days across a 11–50-person team).
Oraion is a Dublin-based AI platform that connects fragmented enterprise data sources and surfaces insights without requiring custom data engineering. Founded in 2024, the company targets mid-market and enterprise buyers who face long setup timelines and hidden costs in traditional data warehousing. The product surface spans data ingestion, permissions and access control, dbt-based analytical modeling, and AI agent-driven task automation. Engineering and sales carry equal hiring weight, indicating a balanced early-stage go-to-market approach.
Oraion's core stack is Python (FastAPI), PostgreSQL, LangChain, LangGraph, LangSmith, dbt, and SQL. Integrations include Slack and Teams. The LLM framework dominance (LangChain, LangGraph, LangSmith, MCP) reflects a platform built around LLM agents and RAG pipelines.
Active projects include scalable RAG pipelines, LLM fine-tuning for enterprise use, permissions and access control layers, data ingestion pipelines, dbt-based analytical modeling, Python/SQL transformations, and AI agent deployment for task automation. Pain points center on production-grade agent deployment, multi-tenant permissions, and latency optimization.
Other companies in the same industry, closest in size