AI engineer in retrieval and agentic systems

Описание:

Kharon operates at the nexus of global security and is on a mission to revolutionize the current landscape. We take complex data as it relates to global security and empower our clients to not only understand the risk associated with their potential business relationships but to operationalize that data so that they can make the best and most informed decisions possible.

Задачи:

Own and advance content discovery in Corestream, improving how customers find, explore, and engage with relevant risk intelligence, while helping shape the technical direction of discovery capabilities across Kharon's products. Redesigning and improving Kharon's hybrid search capabilities, opening new avenues for content discovery - drawing on modern approaches including BM25 tuning, semantic distance, RAG, reranking, and graph-based signals that go beyond traditional semantic similarity. Explore and build new discovery experiences, including agentic search, intelligent content categorization, personalized feeds, and systems that surface relevant connections across large collections of intelligence. Driving the evolution of Kharon's narrative generation pipeline - how we take graph and analytical data and transform them into compelling, accurate written intelligence for customers. Own problems end to end, from identifying opportunities and evaluating approaches to prototyping, implementing, deploying, and continuously improving production systems. Collaborate closely with Product, Data Science, and Engineering, influencing priorities and architectural decisions while maintaining a strong hands-on contribution.

Требования:

A proven interest or experience in global security, financial crimes, sanctions, export controls, or related domains 3+ years of experience in Data Science, AI Engineering, ML Engineering, or related software engineering roles Hands-on experience in information retrieval, including experience with some combination of lexical search, embeddings, vector search, hybrid retrieval, ranking, reranking, and retrieval evaluation Practical experience building LLM-powered applications, RAG pipelines, or agentic workflows. Strong Python skills and the software engineering ability to take an idea from prototype to a reliable, maintainable production implementation Demonstrated initiative and technical ownership - a track record of tackling ambiguous problems independently, challenging assumptions, and driving solutions forward while working with stakeholders Curiosity and a willingness to experiment with emerging approaches while maintaining a pragmatic focus on what actually works Будет плюсом: Experience with graph-aware search or integrating knowledge graph outputs into retrieval systems

Условия:

• £95,000 - £120,000 / year
• £105,000 - £120,000 and discretionary annual bonus
• Monthly 180 days
• Hybrid, 3-days per week
• Fully-sponsored private health insurance
• Pension plan with 3% employer contribution
• Annual leave plus bank holidays

#J-18808-Ljbffr