Staff Applied AI Researcher (Agentic Search)
About this role
Engineers who thrive on solving ambiguous problems at scale and shaping foundational product direction will find a strong fit in this search platform team. Nebius operates a full-stack AI cloud platform and is expanding its R&D efforts to build an agent-native search infrastructure designed specifically for machines rather than human users. Staff and principal researchers tackle complex challenges involving multi-step retrieval, ranking, and real-time reasoning under strict latency constraints.
Daily work centers on designing agent-native retrieval systems, developing methods to ground large language models in dynamic web data, and bridging the gap between theoretical research and production deployment. The group operates within an international environment spanning European research hubs, focusing on high-throughput, low-latency environments that serve thousands of production workloads continuously.
What you will do
- Drive applied research and technical direction across retrieval and ranking systems.
- Design and evolve multi-stage retrieval architectures including query understanding, rewriting, reranking, and iterative retrieval.
- Develop methods for grounding large language models in real-time web data at scale.
- Define and implement new evaluation paradigms and metrics for agentic systems where correctness is not reducible to simple clicks.
- Lead experimentation on modern retrieval approaches such as embeddings, hybrid search, and reranking to bring them into production.
- Analyse trade-offs across relevance, latency, and cost at scale while working closely with engineering teams.
- Own ambiguous problems end to end and mentor engineers to help raise the technical bar of the team.
What you need
- Bring 8 or more years of experience in applied AI, machine learning, or software engineering.
- Demonstrate a proven track record of shipping machine learning or AI systems to production at scale.
- Possess deep experience with search, retrieval, ranking, recommendation systems, or assistants.
- Show a strong understanding of modern deep learning, especially transformers and embeddings.
- Have experience with large language model integrated or knowledge intensive systems.
- Demonstrate experience designing evaluation frameworks and metrics for machine learning systems.
- Maintain strong programming skills in Python alongside at least one language such as Go or C++.
Nice to have
- Experience with large scale search or recommendation systems
- Background in agentic AI systems including agents, tool use, and autonomous workflows
- Experience with RAG, multi step retrieval, or tool use
- Publications, open source, or similar signals of technical depth and impact
Benefits
- Competitive compensation
- Career growth and learning opportunities
- Flexibility and ownership
- Collaborative and innovative culture
- Opportunity to work on impactful AI projects
- International environment and talented teams
Skills and tech stack
Questions candidates ask
What kind of systems does this team build?
The team builds an agent-native search platform, serving as the web access layer for AI systems. This functions as a system designed for machines to actively plan, retrieve, evaluate, and refine information.
What qualifications are required for the role?
Candidates need 8+ years of experience in applied AI, ML, or software engineering, with a proven track record of shipping ML systems to production at scale. Strong Python skills and expertise in deep learning, search, retrieval, or ranking are also required.
What will I actually work on day to day?
Work involves designing agent-native retrieval systems optimized for machine consumption and building systems where LLMs iteratively plan, query, refine, and reason over results. You will also develop ranking approaches for multi-step agent-driven workflows.