About the role
Tavily is building an agent-native search platform designed for AI systems that plan, retrieve, evaluate, and refine information. This Staff Applied AI Researcher will shape research direction and system design for search over real-time web data, with strict latency and reliability requirements. The role owns a specific part of the search stack, such as ranking or iterative retrieval, and brings research advances into production.
The work spans multi-stage retrieval architectures, ranking and retrieval models, LLM grounding, evaluation methods, and trade-offs among relevance, latency, and cost. You will partner with engineering and evaluation teams and mentor researchers and engineers. Applicants should bring 8+ years in applied AI, ML, or software engineering, a record of shipping production systems at scale, deep search or retrieval experience, and top-tier publications. The position is listed as remote in Europe.
Highlights
- Lead applied AI research and technical direction across retrieval and ranking.
- Build and deploy agent-native search systems at web scale.
- Define evaluation metrics for agentic systems beyond click-based measures.
- Mentor teammates and help set the team’s technical bar.
This recap is dataskew's editorial summary, not the company's copy.