AI/ML Engineer, SIGINT & ELINT

About the Role

Auric AI Labs is developing an AI-powered vision intelligence platform for defense applications, building an AI-native intelligence system from scratch. This role focuses on building deep learning models to detect, classify, geolocate, and fingerprint radar and communications emitters directly from raw IQ data.

Key Responsibilities

• Build deep learning models to detect, classify, geolocate, and fingerprint radar and communications emitters directly from raw IQ data.
• Design a synthetic data pipeline.
• Design channel models.
• Design validation methodologies.
• Design tests for simulator accuracy.
• Engage with the DSP world to understand limitations and identify areas for improvement.
• Turn vague objectives into defensible ML problems, defining inputs, outputs, loss functions, and evaluation metrics.

Requirements & Tech Stack

• AI-native instinct: Prioritize learned functions and inductive bias over traditional filters, transforms, or heuristics.
• Ability to reason about model fit to data, understanding the impact of complex-valued signals, long-range temporal dependencies, and distribution shifts.
• Skill in designing experiments that can fail, focusing on falsifying tests before model development.
• Ability to critically read and reproduce research papers, identifying omitted details and underlying mechanisms.
• Proven ability to undertake unassigned work, demonstrated through independent projects or reproduced research.
• Proficiency in translating plain English objectives into specific ML problems with defined inputs, outputs, loss functions, and evaluation metrics.