Data Scientist

Marathon TS is in search of a talented and forward-thinking Computer Vision and Machine Learning Software Engineer with deep expertise in Radio Frequency (RF) systems and signals to join our team. The ideal candidate will be responsible for creating, refining, and deploying sophisticated ML models that merge computer vision with RF signal analysis. This role involves converting complex visual and RF datasets-such as spectrograms, waterfall displays, IQ samples, or multi-sensor data-into meaningful, actionable intelligence for practical deployment. The candidate will integrate traditional signal processing techniques with state-of-the-art deep learning methods to solve complex problems involving image/video recognition and RF waveform classification and interpretation in challenging environments. Core duties include:

Developing, training, fine-tuning, and deploying deep learning and computer vision models for tasks like object detection, segmentation, tracking, categorization, and anomaly detection in both images and video streams. Crafting ML-based solutions specific to RF-related vision applications, such as transforming RF signals (e.g., spectrograms, time-frequency representations, IQ data) into image-like structures suitable for CNNs and other neural network architectures. Building end-to-end ML pipelines, including data collection, preprocessing, feature engineering, model training, evaluation, and deployment-optimized for real-time or edge-based platforms. Implementing sensor fusion techniques by combining visual data with RF signals to enhance detection accuracy, localization, or situational awareness. Performing large-scale data experiments involving RF waveforms, leveraging data augmentation, transfer learning, and model compression to improve performance. Partnering with software developers, RF hardware engineers, data engineers, and subject matter experts to seamlessly integrate ML models into operational systems. Staying current with emerging trends and breakthroughs in computer vision, ML for signal processing, and RF applications; testing and prototyping innovative algorithms. Documenting technical processes, presenting research findings, and contributing to patents or scholarly publications as relevant. Fine-tuning model robustness to perform reliably in real-world conditions, including noisy RF environments and variable imaging scenarios.

Security Clearance: Must hold an active U.S. Security Clearance with eligibility for TS/SCI. Citizenship: U.S. citizenship is mandatory. Educational Background: Bachelor's degree or higher in fields like Computer Science, Electrical Engineering, Data Science, Applied Mathematics, Physics, or comparable disciplines.

Experience: Minimum of three years working hands-on with computer vision and machine learning techniques.

Essential Skills:

Strong grasp of signal processing basics (e.g., Fourier analysis, filtering, modulation schemes) and familiarity with RF tools such as GNU Radio, MATLAB, or proprietary RF datasets. Expert-level programming skills in Python. Experience with deep learning frameworks such as PyTorch and/or TensorFlow/Keras. Proficiency with vision libraries: OpenCV, scikit-image, and related image/video processing methods. Knowledge of ML deployment solutions including ONNX, TensorRT, containerization (Docker), and cloud platforms such as AWS, GCP, or Azure, along with MLOps practices. Robust mathematical foundation in linear algebra, probabilistic models, optimization, and numerical analysis., Demonstrated experience applying ML techniques to RF signals, including RF classification, detection, modulation recognition, or spectrogram analysis using computer vision approaches. Background working with multi-sensor systems or sensor fusion involving RF and visual modalities. Experience in defense, aerospace, autonomous vehicles, spectrum sensing, or wireless communication systems. Familiarity with advanced architectures like transformers, self-supervised learning, or generative models in vision or RF contexts. Knowledge of real-time processing, embedded AI hardware (such as NVIDIA Jetson or FPGA integration), and edge computing solutions. Experience managing extensive datasets and conducting distributed training using GPUs or TPUs.