Auric AI Seeks Research Engineer for Air-Gapped Reasoning and Retrieval System Development
Auric AI is hiring a research engineer to develop a reasoning system designed to operate on millions of messy, multilingual intelligence documents entirely within India, on self-controlled, air-gapped infrastructure. The system will not use fine-tuning or external APIs, relying solely on self-hosted open-weight models. The core challenges involve building retrieval mechanisms that can identify what they missed, reasoning across multiple data sources and hops, and propagating uncertainty through the entire chain of inference. The company emphasizes that a wrong answer delivered with false confidence is worse than no answer. The role requires an individual who can architect, build, and honestly measure solutions for open problems without relying on existing papers or standard playbooks. The ideal candidate should understand language models and retrieval mechanically, not just as APIs, and be able to explain why naive RAG fails on multi-hop temporal questions. 
