AI Scientist – Knowledge Graphs & Memory Systems (AI)

TL;DR

AI Scientist – Knowledge Graphs & Memory Systems (AI): Lead design and development of knowledge layer powering agentic AI systems for representing and retrieving scientific information, technical documentation, and long-term memory of experiments. Focus on scalable knowledge graphs, hybrid retrieval methods, memory consolidation for LLMs, rigorous experimentation, evaluation metrics, and integration into production workflows.

Location: Hybrid in Barcelona, Spain (2/3 days per week onsite at Carrer d'Esteve Terradas 1, Castelldefels)

Company

Building verifiable, interpretable AI systems combining deep learning, formal logic, and physics-based modeling to accelerate semiconductor and photonic hardware development by 30x by 2030.

What you will do

• Lead research on knowledge graphs, retrieval, and memory systems for agentic AI, identifying high-impact directions.
• Design and implement components for knowledge ingestion, representation, indexing, retrieval, and long-term memory with focus on scalability.
• Run experiments to evaluate retrieval and memory quality, define metrics, analyze failures, and iterate improvements.
• Integrate components into agentic workflows for reliable use in research and production.
• Analyze results, document findings, and communicate insights to teams and stakeholders.
• Contribute to publications in leading AI venues.

Requirements

• PhD in AI, ML, CS or related field.
• 2+ years in AI research or applied research engineering with strong technical contributions.
• Hands-on experience building retrieval systems for LLM-based applications or agentic workflows.
• Deep knowledge of retrieval and knowledge representation: knowledge graphs, embedding-based retrieval, hybrids.
• Proficiency in Python for clean, reliable research code.
• Experience with rigorous experiments, metrics, benchmarks, failure analysis.
• Strong collaboration, guiding juniors, communicating complex ideas.

Nice to have

• Graph databases (Neo4j) and knowledge graph tooling.
• Retrieval/indexing systems (Elasticsearch, FAISS, vector DBs), hybrid search.
• RAG, document ingestion, long-context retrieval, agent memory systems.
• Evaluation methods for retrieval/memory, publications or research impact.
• Experience supporting scientific/engineering teams.

Culture & Benefits

• Competitive compensation and stock options.
• Access to cutting-edge tools and collaboration with AI, physics, hardware experts.
• Professional growth: conferences, research presentations, global AI community engagement.
• Impact-driven culture in fast-paced research environment focused on AI-hardware innovation.