About the role
Responsibilities:
- Develop and optimise models for deep text processing on unstructured scientific and legal data (search, classification, entity recognition, and semantic parsing).
- Design and tune robust retrieval-augmented generation systems, experimenting with hybrid search (lexical + vector), chunking strategies, and custom embeddings.
- Evaluate, prompt-engineer, and fine-tune open-source LLMs for extreme domain-specific accuracy.
- Strong experience in NLP, building GenAI products, and fine-tuning.