Principal Scientist - AI/ML Specialization

Role Description

GITI is seeking a Principal Scientist to serve as the senior technical authority on an R&D program focused on passive RF emitter identification and network analysis from real-time sensor data streams. The Principal Scientist leads independent, hands-on analysis of NDF (Network Description File) sensor datasets, provides technical direction across parallel research threads, and serves as the primary technical advisor to the government sponsor. The role spans the full research lifecycle:

• Formulating hypotheses
• Writing and executing analytical code in Python and Jupyter notebooks
• Interpreting and validating results
• Communicating findings to both technical peers and non-specialist stakeholders

This is a deeply technical, hands-on position — the Principal Scientist conducts analysis directly and does not delegate technical work as a substitute for personal proficiency. The candidate will work within a small, distributed team operating in air-gapped Linux environments on resource-constrained tactical edge hardware, with no cloud computing.

Qualifications

• 10+ years of hands-on applied R&D experience in RF systems, signals intelligence, electronic warfare, or related domains.
• Proven ability to quickly acquire domain knowledge; specifically in the areas of wireless digital communications and military techniques, tactics, and procedures.
• Demonstrated ability to independently develop and execute data analyses in Python or equivalent tools on real sensor datasets; must be capable of writing production-quality analytical code, not merely directing others to do so.
• Experience addressing common problems with large quantities of real-world data, such as imputation, noise, bias, and errors.
• Track record of working effectively on constrained-hardware edge systems — no cloud, no discrete GPU — with attention to computational efficiency and multi-core, multi-thread performance on x86 platforms.

Requirements

• Expert-level career professional recognized as a technical authority in RF systems, signals intelligence, or a closely related applied domain.
• Exercises broad independent judgment in defining research approach, evaluating methods, and interpreting results.
• Operates with minimal supervision; accountable for the scientific integrity and practical relevance of program research outputs.
• Advanced degree (MS or PhD) with 10+ years of hands-on applied R&D experience.

Desired Skills

• Deep familiarity with RF signal characteristics, sensor phenomenology, and the interpretation of passive receiver data — including recognition of processing artifacts, attribution ambiguities, and the limits of sensor-derived measurements.
• Hands-on experience applying machine learning — particularly metric learning, deep learning networks, or similarity-learning architectures — to RF or time-series signal data, including feature engineering, training pipeline development, and model validation.
• Familiarity with TDMA network protocols, emitter identification techniques (CID/PID), and the signal processing challenges of dense, contested electromagnetic environments.
• Experience with interferometric direction-finding, TDOA geolocation, or related passive geolocation methods, including practical knowledge of their failure modes and accuracy limitations.
• Experience with binary serialization formats (FlatBuffers, Protocol Buffers) and high-throughput sensor data pipelines operating in near-real-time on resource-constrained hardware.
• Background in statistical signal processing — error ellipses, bearing estimation uncertainty, feature reliability under noise — with the ability to distinguish statistically significant findings from artifacts of small sample size or improper normalization.

Relevant Certifications

• Professional certifications in data science, signal processing, or related technical fields.
• Advanced academic credentials (PhD, MS) in a relevant quantitative discipline are strongly preferred and may substitute for certifications.