Responsibilities

• Research, develop, and improve the machine learning models that extract semantic signals from structured and unstructured enterprise data sources.
• Apply and advance unsupervised learning techniques including pattern mining, hierarchical clustering, and graph-based methods — to surface meaningful business context at scale.
• Develop NLP and embedding-based approaches to ground unstructured knowledge (documentation, collaboration tools, BI metadata) into the Context Graph.
• Build validation, scoring, and drift detection frameworks to ensure the semantic model stays accurate and current as organizational behavior evolves.
• Collaborate with engineering to bring models into production within an agentic, multi-step reasoning architecture.
• Stay close to the research frontier across semantic AI, agentic systems, and GenAI, and bring relevant advances into the product.

Requirements

• 5+ years of hands-on data science experience shipping production models.
• Strong foundation in unsupervised learning – clustering, pattern mining, graph algorithms.
• Solid experience with NLP pipelines: embeddings, entity extraction, semantic similarity, topic modeling.
• Hands-on experience with large language models, prompt engineering, RAG architectures, and grounding techniques.
• Hands-on experience with agentic AI systems – multi-step reasoning, tool use, agent feedback loops, and context window management.
• Proficiency in Python and the ML/data science stack.
• Comfort with SQL and reasoning about complex query patterns, this is some of the data you’ll be working with.

Nice to Have

• Experience with knowledge graphs or ontology modeling.
• Familiarity with enterprise data infrastructure – warehouses, BI tools, semantic layers.
• Background in statistical drift detection or temporal modeling.
• Experience building or evaluating AI agents in enterprise settings.