Principal AI Researcher, Retrieval & Agentic Systems

Nebius; Greater London; On-site; GBP 120,000 - 180,000; Full time

Benefits offered by this job

Competitive compensation

Career growth & learning

Flexibility & ownership

Collaborative culture

Impactful AI projects

International environment

Job summary

Nebius is seeking a Staff or Principal Applied AI Researcher to advance agent native retrieval systems for real-time web data. You will own research directions, design scalable architectures, and ship measurable AI impact in production.

The role emphasizes building systems where LLMs plan, query, refine, and reason over data with strict latency constraints, collaborating across engineering and product teams.

Qualifications

• 8+ years of experience in applied AI, ML, or software engineering.
• Proven track record of shipping ML/AI systems to production at scale.
• Deep experience with search, retrieval, ranking, or assistant systems.
• Strong Python skills and knowledge of Go/C++ or similar.
• Experience with LLM-integrated or knowledge-intensive systems.

Responsibilities

• Drive applied research and technical direction across retrieval and ranking systems.
• Design multi-stage retrieval architectures (query understanding, rewriting, reranking).
• Ground LLMs in real-time web data at scale and evaluate new paradigms.
• Lead experimentation on embeddings, hybrid search, and reranking in production.
• Analyze trade-offs across relevance, latency and cost at scale.

Skills

8+ years experience

Production ML systems

Search/Retrieval systems

Python programming

Go or C++

LLM integrated systems

Evaluation frameworks for ML

Job description

Nebius is seeking a Staff or Principal Applied AI Researcher to advance agent native retrieval systems for real-time web data. You will own research directions, design scalable architectures, and ship measurable AI impact in production.

The role emphasizes building systems where LLMs plan, query, refine, and reason over data with strict latency constraints, collaborating across engineering and product teams.