Gute Arbeit beginnt mit dir.

• Design and train models from raw I/Q recordings for detection, classification, and related tasks
• Research and implement machine-learning methods from other domains to improve data pipelines
• Define evaluation test sets, metrics, and scenarios reflecting domain shifts between sensors, sites, and interference environments
• Develop compact, deployable expert models for edge hardware
• Continuously test pipelines in real-world conditions and use feedback to improve them
• Integrate new data sources into the data platform
• Set the direction for RF machine learning at Datacept
• Make architectural decisions for the RFML foundation
• Work directly with the founders and build the team as the company grows
• Stay connected to hardware, founders, and customers at protected sites

Requirements

• Experience building and shipping ML systems that people depend on, preferably on signal-like data such as audio, time series, sensor streams, images, video, or RF
• Strong understanding of self-supervised learning and ability to explain why methods work
• Strong mathematical fundamentals and comfort formulating problems before solving them
• Ability to go from idea to prototype to deployed model independently
• Ability to work with few fixed structures and changing priorities
• Preparedness for the intensity of a founding role, including long and unconventional working hours, field trials, and deployments
• Willingness to contribute to Europe's technological sovereignty
• Helpful but not required: prior RF or signal-centric ML experience, including spectrum sensing, modulation recognition, SIGINT, or EW
• Helpful but not required: self-built software/hardware projects
• Helpful but not required: signal processing basics, sampling, spectral analysis, I/Q representation, SDR, communications engineering, or embedded background
• Helpful but not required: publications, open-source work, or production architectures

Core Competencies

Demonstrates expertise in designing and deploying machine learning models for RF and signal-like data, with a strong foundation in self-supervised learning and mathematical problem formulation. Capable of integrating new data sources and making architectural decisions to enhance data pipelines and model performance.

Highest-signal resume keywords

• Machine Learning Systems Development
• Self-Supervised Learning
• Signal Processing
• Architectural Decision-Making
• Prototype to Deployed Model Transition

Hard Skills

• Machine Learning
• Model Training
• Data Pipeline Improvement
• Mathematical Fundamentals
• Signal Processing Basics
• I/Q Representation
• Spectrum Sensing
• Modulation Recognition
• Field Trials
• Deployment

Soft Skills

• Adaptability
• Team Building
• Communication
• Problem Solving
• Independence

Industry Keywords

• RF Machine Learning
• Technological Sovereignty
• Signal-Centric ML
• Field Deployments
• Unconventional Working Hours

Tools & Technologies

• Edge Hardware
• Data Platform
• SDR
• Embedded Systems