RF Signals and Data Analyst

About the role

Quartermaster AI is seeking an experienced RF Signals Analyst to lead our signal characterization and data labeling efforts. This role focuses on turning real-world RF sensor data into structured ground truth for machine learning.

Responsibilities

• Analyze RF event data using IQ derived representations such as spectrograms, waterfall views, PSDs, and metadata to identify, classify, and tag signals of interest.
• Help define and maintain a scalable maritime RF labeling taxonomy, including signal classes, confidence levels, rejection categories, and ambiguity handling.
• Build and refine high-quality labeled datasets for machine learning, ensuring labels are technically defensible, consistent, and auditable.
• Identify and document recurring host vessel interference, platform artifacts, and environmental noise to support rejection library development.
• Collaborate with DSP and ML engineers to review false positives, false negatives, and edge cases, and improve labeling standards over time.
• Use available contextual data such as AIS, camera imagery, collection metadata, and sensor state to support signal interpretation when appropriate.

Qualifications

• 3+ years of experience in one or more of the following: RF signal analysis, SDR-based signal review, EW/SIGINT/ELINT analysis, RF dataset creation, or technical signal characterization.
• Practical experience working with RF data products such as IQ captures, spectrograms, waterfall plots, PSDs, or other time-frequency representations.
• Experience working with structured labeling, annotation, classification, or technical review workflows where consistency and traceability matter.
• Comfort working in a Linux-based environment using Python, SDR tools, notebooks, or other RF analysis environments to inspect, organize, and process signal data.
• Ability to communicate clearly with engineers and translate signal observations into actionable labeling guidance.
• Experience in maritime RF environments or other cluttered, interference-heavy operational environments.
• Understanding of how label quality, taxonomy design, multi-sensor context (e.g., AIS, EO/IR, or geolocation), and rejection categories affect downstream ML training and evaluation.
• Active clearance or ability to obtain and maintain a Secret clearance.