outmarket - Applied AI Scientist (PhD)

Remote - USA+ Equity1mo ago

Remote NA Insurance Artificial Intelligence Applied Scientist Python Hugging Face Transformers B2B

Requirements

• PhD in Computer Science, Machine Learning, NLP, or a related field. • Strong research background with publications in top-tier AI venues such as NeurIPS, ACL, ICML, or EMNLP. • Hands-on experience with LLMs, transformers, embeddings, or neural information retrieval. • Proficiency in Python and ML tooling such as PyTorch, Hugging Face, and LangChain. • Track record of applying research in practical, production-oriented systems. • Strong technical judgment about what is novel, what is useful, and what is actually ready to ship. • Experience with unstructured data such as PDFs, forms, contracts, or portals. • Background in enterprise or B2B AI applications. • Familiarity with insurance, legal, or document-heavy industries. • WHAT YOU’LL GET • A high-impact role on a team that moves quickly and builds with purpose. • Competitive compensation and meaningful equity. • Remote-first flexibility. • Access to real data, real users, and fast feedback cycles.

Responsibilities

• Research and develop LLM-based solutions for NLP, document intelligence, semantic search, and data extraction. • Design and improve prompt strategies, retrieval-augmented generation systems, and fine-tuned models. • Partner with product and engineering teams to bring AI capabilities into customer-facing workflows. • Build evaluation pipelines and benchmarks for accuracy, performance, and robustness. • Stay current on new research and rapidly test promising techniques in practical settings. • Operate as both a scientist and builder, with direct responsibility for whether ideas survive contact with real data.

Benefits

• Apply cutting-edge AI research to live customer workflows and production systems. • Work directly with founders and a highly experienced technical team. • Build solutions for document intelligence, quote comparison, workflow optimization, and other high-value use cases. • See your work move from research to deployment in weeks, not quarters. • Help define how applied AI creates advantage in a large, underserved market.