Threat Intel Data Scientist

About the role

Deloitte’s Global Centralized Cybersecurity Threat Intelligence team is looking for a Data Scientist to design AI‑driven solutions that automate the processing, normalization, enrichment, and analysis of threat intelligence data. You will work with large‑scale structured and unstructured datasets to enhance cybersecurity intelligence products used worldwide.

Key responsibilities

• Design, develop and continuously improve AI and Machine Learning models that support cyber analysts and intelligence‑driven decision making.
• Transform, normalize and enrich data from multiple internal and external sources into the Threat Intelligence Platform (TIP) and related tools.
• Apply Generative AI technologies (LLMs, RAG, embeddings, semantic search, prompt engineering) to automate and enrich threat‑intelligence workflows.
• Analyze large datasets to identify patterns, anomalies and emerging threats using advanced analytics, correlation and clustering techniques.
• Identify and integrate new intelligence data sources to broaden analytical capabilities.
• Create and maintain technical documentation and SOPs for scalable, well‑governed solutions.

Required profile

• 4+ years of experience in Data Science, Machine Learning, Analytics, Data Engineering or a related technical field.
• Strong background in Python and SQL, with experience building APIs and data processing pipelines.
• Hands‑on experience developing and deploying ML/AI solutions and a solid understanding of statistical modeling and large‑scale data analysis.
• Experience with Generative AI technologies such as LLMs, prompt engineering, RAG, embeddings or semantic search.
• Ability to work with both structured and unstructured data and communicate analytical insights effectively.
• Prior cybersecurity experience is a plus but not mandatory.

Required skills

• Python
• SQL
• APIs
• Data processing pipelines
• Machine Learning
• Artificial Intelligence
• Generative AI
• Large language models (LLMs)
• Prompt engineering
• Retrieval‑augmented generation (RAG)
• Embeddings
• Semantic search
• Statistical modeling
• Large‑scale data analysis