Source description
About the role
As an engineer at Auric AI Labs, your primary responsibility will be to build deep learning models that detect, classify, geolocate, and fingerprint radar and communications emitters directly from raw IQ data. We are developing an AI-native intelligence system that operates on raw IQ data with minimal pre-processing, learning its own features, and replacing the deterministic stack rather than supplementing it. Our goal is to achieve a 5-10x improvement in every metric that matters, such as detection in low SNR, classification of unknown waveforms, specific emitter ID, real-time anomaly detection, and cognitive EW. The challenges you will face include signals deliberately designed to defeat traditional methods, lack of real training data for crucial signals, uniqueness of each emitter requiring open-set, fine-grained, unlabelled approaches, and skepticism from the DSP world about the feasibility of our approach. You will need to design a synthetic data pipeline, channel models, validation methodology, and tests to ensure the reliability of your models in real-world scenarios. We are looking for someone with specific characteristics rather than a generic list of traits: - You possess an AI-native instinct, thinking in terms of data, loss, and architecture rather than relying on hand-engineered features. - You can reason about whether a model fits its data and understand the implications of complex-valued signals, long-range temporal dependencies, and small distribution shifts. - You design experiments that can fail, focusing on tests that could potentially prove your assumptions wrong. - You have a keen eye for details in research papers, identifying hidden information that authors may have omitted. - Your best work is self-driven, showing initiative and passion for solving challenging problems beyond assigned tasks. - You excel at translating vague objectives into well-defined ML problems, understanding the inputs, outputs, loss functions, and evaluation criteria required for success. If you are someone who thrives on pushing the boundaries of AI technology, questioning conventional approaches, and solving complex problems with innovative solutions, we encourage you to apply for this role at Auric AI Labs. As an engineer at Auric AI Labs, your primary responsibility will be to build deep learning models that detect, classify, geolocate, and fingerprint radar and communications emitters directly from raw IQ data. We are developing an AI-native intelligence system that operates on raw IQ data with minimal pre-processing, learning its own features, and replacing the deterministic stack rather than supplementing it. Our goal is to achieve a 5-10x improvement in every metric that matters, such as detection in low SNR, classification of unknown waveforms, specific emitter ID, real-time anomaly detection, and cognitive EW. The challenges you will face include signals deliberately designed to defeat traditional methods, lack of real training data for crucial signals, uniqueness of each emitter requiring open-set, fine-grained, unlabelled approaches, and skepticism from the DSP world about the feasibility of our approach. You will need to design a synthetic data pipeline, channel models, validation methodology, and tests to ensure the reliability of your models in real-world scenarios. We are looking for someone with specific characteristics rather than a generic list of traits: - You possess an AI-native instinct, thinking in terms of data, loss, and architecture rather than relying on hand-engineered features. - You can reason about whether a model fits its data and understand the implications of complex-valued signals, long-range temporal dependencies, and small distribution shifts. - You design experiments that can fail, focusing on tests that could potentially prove your assumptions wrong. - You have a keen eye for details in research papers, identifying hidden information that authors may have omitted. - Your best work is self-driven, showing initiative and passion for solving challenging problems beyond assigned tasks. - You excel at translating vague objectives into well-defined ML problems, understanding the inputs, outputs, loss functions, and evaluation criteria required for success. If you are someone who thrives on pushing the boundaries of AI technology, questioning conventional approaches, and solving complex problems with innovative solutions, we encourage you to apply for this role at Auric AI Labs.