About the job
At Sentra, we are pioneering the development of organizational superintelligence through innovative memory infrastructure that intelligently processes time, causality, and context. As a Machine Learning Research Scientist, you will address fundamental challenges in knowledge representation, temporal reasoning, and semantic compression. Your mission will be to design and implement sophisticated systems that preserve the execution state for entire organizations, transforming millions of micro-events into robust knowledge and identifying patterns for predicting future events.
Key Responsibilities
Develop LLM-powered information extraction pipelines to convert unstructured communications and textual data into structured entity-relationship models.
Create memory consolidation algorithms that validate information through multiple observations, merge duplicate entities, and efficiently prune transient data.
Architect temporal knowledge graphs that represent organizational execution states as dynamic, continuously updated frameworks instead of static records.
Implement graph attention mechanisms and reasoning systems for intricate causal queries regarding blockers, dependencies, and outcome patterns.
Conduct research on lossy semantic compression using information-theoretic principles to distill event streams into query-relevant long-term memory.
Design entity resolution systems that effectively manage identity evolution, where entities may merge, split, and transform over time.
Construct meta-learning systems that uncover organizational patterns and discern when current situations align with historical indicators of success or failure.
Innovate privacy-preserving cross-organizational learning approaches utilizing federated learning and differential privacy techniques.
Publish research findings and actively contribute to the wider research community focused on knowledge graphs and organizational intelligence.
